-
tokenizers
Fastest implementation of the most used tokenizers in AI
-
fastembed
generating vector embeddings, reranking locally
-
genai
Multi-AI Providers Library for Rust. (OpenAI, Gemini, Anthropic, DeepSeek, Ollama, OMLX, AWS Bedrock, Vertex, Xai, Groq, Kimi, GLM and many more)
-
llama-cpp-2
llama.cpp bindings for Rust
-
zerostack
Minimalistic coding agent written in Rust, optimized for memory footprint and performance
-
smg-mcp
Model Context Protocol (MCP) client implementation
-
rig
An opinionated library for building LLM powered applications
-
scxtop
sched_ext scheduler tool for observability
-
mecha-cli
The mecha CLI: an agent harness for local models
-
onnx-ir
pure Rust library for parsing ONNX models into an intermediate representation that can be used to generate code for various ML/DL frameworks
-
llm
unifying multiple LLM backends
-
dynamo-tokens
Token management tools
-
smg-grpc-client
gRPC clients for vLLM, TensorRT-LLM, MLX, TokenSpeed, and SGLang backends
-
ruvector-core
High-performance Rust vector database core with HNSW indexing
-
dynamo-renderer
Standalone OpenAI chat-template / prompt formatting (HF chat_template via minijinja)
-
adk-model
LLM model integrations for Rust Agent Development Kit (ADK-Rust) (Gemini, OpenAI, Claude, DeepSeek, etc.)
-
code2prompt
Command-line interface for code2prompt
-
claude-wrapper
A type-safe Claude Code CLI wrapper for Rust
-
clawshell
A security privileged process for the OpenClaw ecosystem
-
agentic_ssh
A minimalist, secure engineering primitive for agentic SSH execution and detached background operations
-
hf-xet
Client library and tooling for the Hugging Face Xet data storage system
-
ilmari
Minimal tmux popup radar for coding agents
-
looop
A tiny, portable, Kubernetes-shaped control loop for your work
-
hedos
Discover, install, and serve the local models already on your machine
-
think-tool
CLI for coordinating persistent agent sessions on complex projects
-
ai
LLM library for Rust with streaming, tool calling, OAuth helpers, and a lightweight agent loop
-
atim
AI Agent Through IM
-
droidsight
MCP server that drives a real Android device over ADB and attaches the resulting screen to every action, so an agent never needs a separate screenshot call. Single native binary, with…
-
lowfat
Token-aware output and file content filter for LLM environments
-
deepwiki-rs
deepwiki-rs(also known as Litho) is a high-performance automatic generation engine for C4 architecture documentation, developed using Rust. It can intelligently analyze project structures…
-
ai-usagebar
Omarchy/Waybar widgets + TUI for tracking multi-provider AI plan usage
-
foundry-local-sdk
Local AI model inference powered by the Foundry Local Core engine
-
llm_adapter
adapting various language model APIs to a unified interface
-
openai_dive
OpenAI Dive is an unofficial async Rust library that allows you to interact with the OpenAI API
-
openai-protocol
OpenAI-compatible API protocol definitions and types
-
llama-cpp-4
llama.cpp bindings for Rust
-
embed_anything
Embed anything at lightning speed
-
obsidian-rs-mcp
An MCP server for interacting with Obsidian vaults
-
hrdr
hrdr — herder: a fast, agentic coding harness for OpenAI-compatible models
-
ct2rs
Rust bindings for OpenNMT/CTranslate2
-
swiftide
Composable LLM agents, typed task graphs, and streaming RAG pipelines in Rust
-
autoagents
Agent Framework for Building Autonomous Agents
-
openai-harmony
OpenAI's response format for its open-weight model series gpt-oss
-
shaic
Sync AI-agent skills, rules, commands, and MCP servers via git
-
sema-lsp
Language Server Protocol server for Sema
-
terraphim_service
Terraphim service for handling user requests and responses
-
bamboo-domain
Domain models and shared types for the Bamboo agent framework
-
mcp-hub
Tools-only Model Context Protocol aggregation server
-
yolop
terminal coding agent built on everruns-core
-
transcribe-rs
help you transcribe audio
-
thoughts-bin
CLI for flexible thought management using filesystem mounts
-
rsigma
CLI for parsing, validating, linting and evaluating Sigma detection rules
-
deepseek-tui
DEPRECATED — renamed to codewhale. Install
codewhaleinstead (binary: codewhale). -
anda_brain
🧠 Anda Brain (大脑) — Autonomous Graph Memory for AI Agents
-
greplm-cli
greplm command-line interface
-
llm_models_spider
Auto-updated registry of LLM model capabilities (vision, audio, etc.)
-
crw-extract
HTML extraction and markdown conversion engine for the CRW web scraper
-
harn-builtin-meta
Shared type definitions for Harn builtin signatures (BuiltinSignature, Param, Ty). Consumed by both harn-parser and harn-vm so signatures stay cycle-free.
-
brokk-anvil
Anvil: Rust ACP server with first-run setup for Codex, Ollama, and OpenRouter
-
supercode-reduce
Optional lossless, reversible session reduction for Volter Harness
-
edda-bridge-claude
Claude Code hooks, transcript ingest, and context injection for Edda
-
lean-ctx
LeanCTX | OSS: the open-source runtime for the Context Gateway for AI agents. CLI, MCP server, and supported proxy integrations.
-
repograph
Register, group, and expose local git repos as structured context for AI agents
-
symbiont
Agent harness for hot-reloadable function evolution in Rust
-
stakpak
Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
create-django-bolt
Ultra-fast CLI tool to scaffold production-ready Django-Bolt applications with Docker, Kubernetes, and Fly.io
-
qqq
Quick Query Quorum: an in-memory multi-user chat with an AI facilitator
-
instructors
Type-safe structured output extraction from LLMs. The Rust instructor.
-
vct-core
Vibe Coding Tracker core library - parse local AI coding assistant session data into CodeAnalysis results
-
solverforge-cli
CLI for scaffolding and managing SolverForge constraint solver projects
-
codewandler-flux-policy
flux pure authorization: default-deny grant evaluation with trust levels, scopes, and wildcard/path matching
-
asimov-server
ASIMOV Software Development Kit (SDK) for Rust
-
cloudllm
A batteries-included Rust toolkit for building intelligent agents with LLM integration, multi-protocol tool support, multi-agent orchestration, and MentisDB-backed durable memory
-
cgip
Terminal client for interacting with Chat GPT that allows you to build and manipulate contexts
-
dynamo-llm
Dynamo LLM Library
-
aether-llm
Multi-provider LLM abstraction layer for the Aether AI agent framework
-
mcpstead
MCP Gateway
-
hallouminate-domain
Domain layer for hallouminate
-
teach-cli
A teaching harness: turn any directory into a dynamic, student-led course
-
turbovault
Production-grade MCP server for Obsidian vault management - Transform your vault into an intelligent knowledge system for AI agents
-
embacle
LLM runner library — 12 AI CLI tools, ACP, and HTTP APIs (Anthropic, Gemini, Cohere, Groq, OpenRouter, OpenAI-compatible) as pluggable LLM providers with fallback chains, agent loop, guardrails, and cost tracking
-
git-ai
Git AI: Automates commit messages using ChatGPT. Stage your files, and Git AI generates the messages.
-
tandem-enterprise-contract
Public enterprise contract and tenant context types for Tandem
-
whathaveidone
CLI tool for daily standups with Git and Ratatui
-
agentic-git
A guarded, transparent git for AI coding agents — a PATH shim over the real git with per-agent worktree routing, a deny guardrail matrix, commit provenance, and audited bypass
-
harn-ir
CFG and invariant analysis for the Harn programming language
-
zeph-subagent
Subagent management: spawning, grants, transcripts, and lifecycle hooks for Zeph
-
ferrisgrid-cli
Terminal-first visual control CLI for local AI agents
-
tokf-filter
Pure filter engine for tokf — config-driven output compression
-
embsearch-cli
CLI + stdio daemon for embsearch: index, query, update, and serve embeddings
-
apl-cmf
APL ↔ CPEX bridge — extension → AttributeBag mapping
-
edgee-compressor
HTTP response compression library for Edgee
-
logzip
Compress logs for LLM analysis — CLI + MCP server
-
linguafranca
LLM API format converter — convert between OpenAI, Anthropic, and Open Responses formats
-
shipmates
A crew of specialist AI agents and command workflows that drive a GitHub issue to a reviewed, CI-green pull request
-
musicgpt
Generate music based on natural language prompts using LLMs running locally
-
axllm
Generated Ax runtime library
-
atlasctl-protocol
Wire types shared by the atlasctl agent, the CLI, and the browser client
-
mcpr-proxy
A proxy for MCP Apps/Servers — routes JSON-RPC, observes traffic, authenticates, and secures MCP
-
algocline
LLM amplification engine — MCP server with Lua scripting
-
yomo
A QUIC-based runtime for AI-LLM tool routing and serverless execution
-
llm_json
fix malformed JSON outputs from LLMs
-
converge-provider
The capability and backend selection contract for Converge providers
-
semantex-cli
CLI for semantex semantic code search
-
commandok
Spotlight-like command generator for your terminal, powered by LLMs
-
tokn-convert
Request and response conversion pipeline across tokn endpoint formats
-
molo
An embeddable Rust agent runtime and harness framework for building safe, extensible tool-calling agents
-
famulus
LSP server integrating LLMs
-
siumai
Ergonomic, provider-faithful AI model interfaces for Rust
-
m1nd-core
Core graph engine and reasoning primitives for m1nd
-
ai-agents-hitl
Human-in-the-loop support for AI Agents framework
-
agentx-adk-model
LLM model integrations for Rust Agent Development Kit (ADK-Rust) (Gemini, OpenAI, Claude, DeepSeek, etc.)
-
motosan-ai
Rust SDK for multi-provider AI chat
-
stillo
AI-native terminal browser
-
assay-runner-schema
Internal/experimental substrate for Assay measured-run workflows. Versioned schema types and constants for the Assay-Runner v0 contracts (Phase 2D Slice 1). No standalone product guarantee;…
-
otelite-tui
Terminal User Interface for Otelite OpenTelemetry receiver
-
api_huggingface
HuggingFace's API for accessing large language models (LLMs) and embeddings
-
open-kioku-resolution
Deterministic symbol and relationship resolution for Open Kioku code intelligence
-
pizen-lynx-mcp
Internal Model Context Protocol implementation for Lynx (canonical entrypoint:
lx mcp) -
battery-pack
Curated crate bundles with docs, templates, and agentic skills
-
diffler
Standalone magit-style git TUI and code reviewer, on its own or with your coding agent
-
skrills_sync
Preference synchronization and validation sync utilities for skrills
-
turbocommit
A CLI tool to create commit messages with OpenAI GPT-6 for Git and Jujutsu repositories
-
infigraph-mcp
MCP server for infigraph — exposes code graph tools to AI agents
-
smix-capsule
Runner lifecycle for smix: bring an XCUITest runner up on a simulator, watch it, take it down
-
ctxctl
CLI-first, stateless context layer for AI coding agents (optional MCP adapter)
-
elevenlabs_rs
A lib crate for ElevenLabs
-
skillfile-sources
Source fetching and caching for skillfile: resolver, strategies, sync
-
raphtory
temporal graph library
-
cuca
Compact Universal Client for Agents: an asynchronous LLM client that parses multi-modal SSE streams into typed blocks behind one unified request/response abstraction
-
ascent-research
— an incremental research workflow CLI for AI agents. Every session resumes; knowledge accretes across runs. Mixes HTTP, browser, and local file ingest into a durable per-session wiki + figure-rich HTML report.
-
uteke-core
The Brain for Your AI — persistent memory and smart context management
-
papers-cli
CLI for academic paper search, management, and local RAG
-
vs-engine-webkit
Browser engine bindings for vibesurfer
-
lineprior
Explainable action ranking from historical sequences, with confidence, context backoff, evaluation, and tuning
-
turul-http-mcp-server
HTTP transport layer for Model Context Protocol (MCP) servers
-
journal-mcp
MCP server (and maintenance CLI) for the journal library — project canonical history
-
bitrouter-observe
BitRouter observability plugin: OTLP traces and metrics with multi-tenant attribution
-
cerememory-cli
Command-line interface for Cerememory
-
rskim
The most intelligent context optimization engine for coding agents. Code-aware AST parsing, command rewriting, output compression.
-
arai
AI coding rules that actually work. Enforce instruction files via hooks — CLAUDE.md, .cursorrules, copilot-instructions, and more.
-
open-kioku-impact
Impact analysis for code changes in Open Kioku
-
lattice-inference
Pure Rust transformer inference engine — safetensors loading, SIMD matmul, BGE/Qwen3 embeddings
-
imorgon-mcp
Model Context Protocol server for imorgon
-
oxios-memory
Tiered agent memory — extracted from oxios-kernel per RFC-018
-
zerobox-sandboxing
Sandbox any command with file, network, and credential controls
-
claudy
Modern multi-provider launcher for Claude CLI
-
agentkit-mcp
Model Context Protocol integration for agentkit, including transports, auth, and adapters
-
codewhale-protocol
App-server protocol frames for Codewhale runtime integrations
-
shore-cli
Command-line client (
shore) for the Silvershore daemon -
blazen-controlplane
Distributed workflow control plane for Blazen — workers connect inbound to a central server
-
actually
forces Claude Code agents into creative thinking by pitting them against each other as contrarian strategists
-
dirge-agent
Minimalistic coding agent written in Rust, optimized for memory footprint and performance
-
trusty-common
Shared utilities and provider-agnostic streaming chat (ChatProvider, OllamaProvider, OpenRouter, tool-use) for trusty-* projects
-
mimi-ai
A basic AI agent harness
-
greplm-mcp
greplm Model Context Protocol (MCP) stdio server
-
sacp-conductor
Conductor for orchestrating SACP proxy chains
-
kalosm-sample
A common interface for token sampling and helpers for structered llm sampling
-
pensyve-cli
Pensyve command-line interface — universal memory runtime for AI agents
-
tysm
Batteries-included Rust OpenAI Client
-
miasma
Trap malicious AI web scrapers in an endless pit of poisoned training data
-
ccbar
Fast, configurable statusline for Claude Code
-
robit-agent
Agent runtime, tool system, skill system, and frontend trait for robit
-
routa-core
Routa.js core domain — models, stores, protocols, and JSON-RPC (transport-agnostic)
-
raysense
Architectural X-ray for your codebase. Live, local, agent-ready.
-
oven-llm
calling LLM providers with a unified API
-
openai
An unofficial Rust library for the OpenAI API
-
cv-llm
LLM-backed features over clustervision sessions (distillation) via OpenRouter / Anthropic
-
zapmyco
An AI-powered CLI tool for interactive chat with LLMs
-
opencrabs
The autonomous, self-improving AI agent. Single Rust binary. Every channel. Recommended: the 40MB prebuilt binary for macOS, Linux and Windows: https://github.com/opencrabs/opencrabs/releases
-
polyforge-toolrunner
Allowlisted, typed-argument, no-shell command runner producing verified tool attestations
-
ares-types
Core types and tenant models for ARES
-
shiotsuchi-mcp
MCP server for Shiotsuchi Search — Claude Desktop integration
-
agent-sdk
Rust Agent SDK for building LLM agents
-
tooned
CLI: convert/check/pipe/wrap, the .tooned/ index, agent hook installers, and the MCP server
-
anyllm_providers
Provider and model catalog for anyllm-proxy
-
agentic-outer-dag-bin
External outer-DAG driver for worktree→PR→CodeRabbit loops
-
muthr
Runtime management CLI
-
lellm-graph
Graph/Node/Edge orchestration layer for LeLLM — with State, Delta, Checkpoint
-
chromadb
client library for the ChromaDB vector database
-
mcpmesh
Share MCP servers with people you trust — peer to peer, default-deny, no accounts
-
hrdr-agent
hrdr agent loop — drives an OpenAI-compatible model through tool calls until a coding task is done
-
symforge
— in-memory code intelligence MCP server for Claude Code, Codex, and Gemini CLI
-
outlines-core
Structured Generation
-
introspection-sdk
Rust SDK for Introspection — continuously improve your AI systems with production feedback and frontier practices
-
astrid-events
Event bus for Astrid secure agent runtime
-
codewandler-flux-provider
flux provider abstraction: the Provider trait, request/stream types, model resolution
-
oneiros
A sqlite-based cognition kit for long-term agent language models
-
ollmo
OpenAI API-compatible LLM gateway for personal use
-
turbovault-export
Reporting and export utilities for TurboVault Server
-
aimx
SMTP for AI agents. No middleman. Self-hosted mail server (AI Mail Exchange) with built-in MCP.
-
anno
NER, coreference resolution, relation extraction, PII detection, and zero-shot entity types
-
monothink
A totally non-agentic coding assistant
-
prodex
OpenAI profile pooling and safe auto-rotate for Codex CLI and Claude Code
-
mockql-cli
Command-line interface for GenAI-powered GraphQL response mocking via @mock
-
rig-core
An opinionated library for building LLM powered applications
-
llama-herd
A native Rust TUI and CLI companion for orchestrating, pairing, and routing local LLM services driven by llama.cpp's llama-server
-
localcoder
oxink
-
astraea-rag
GraphRAG building blocks for AstraeaDB: subgraph extraction and linearization of graph context for LLM prompts
-
openai-client-base
Auto-generated Rust client for the OpenAI API
-
fluers-core
Core agent primitives and model abstractions for Fluers (port of pi-agent-core + pi-ai)
-
skippy-cache
Cache primitives for Skippy staged runtime serving
-
hai-cli
A CLI with a REPL for hackers using LLMs
-
cats
Coding Agent ToolS - A comprehensive toolkit for building AI-powered coding agents
-
apcore-mcp
MCP bridge for apcore — expose apcore modules as MCP tools
-
rigorix-mcp
MCP Gateway server for Rigorix — protocol types, transports, session management, tool routing, resource/prompt providers
-
cageforge-windows
Windows-native sandbox backend for Rust applications
-
xet-client
Client library for communicating with Hugging Face Xet storage servers. Use through the hf-xet crate.
-
virustotal-rs
Rust SDK for VirusTotal API v3
-
ironclaw
Secure personal AI assistant that protects your data and expands its capabilities on the fly
-
noveum-ai-gateway
A high-performance AI Gateway proxy for routing requests to various AI providers, offering seamless integration and management of multiple AI providers
-
cartog-rag
Semantic search and RAG pipeline for cartog
-
kaish-help
Composable help & instructions content for kaish — shared by the kernel, REPL, MCP server, and embedders
-
guardrail
Defensive guardrails for AI coding agents — block destructive commands via hooks
-
aggit
S3-backed, git-versioned object storage for agents
-
llmshim
Blazing fast LLM API translation layer in pure Rust
-
fanin-mcp
A standalone, stdio-native MCP proxy that federates many upstream MCP servers behind a single endpoint
-
cooklang-import
importing recipes into Cooklang format
-
regenerator2000
An interactive disassembler for the CPU 6502, focused mostly on Commodore 8-bit computers
-
cargo-ai
Build lightweight AI agents with Cargo. Powered by Rust. Declared in JSON.
-
omni-dev
AI-powered git commit rewriter, PR generator, and MCP server for Jira, Confluence, and Datadog
-
gline-rs
Inference engine for GLiNER models
-
terraphim_types
Core types crate for Terraphim AI
-
hook0-mcp
MCP Server for Hook0 Open-Source Webhooks as a Service
-
oxidoc
documentation engine CLI — oxidoc dev & oxidoc build
-
ag-forge
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
trackforge
A unified, high-performance computer vision tracking library
-
connector-mcp-server
MCP server for tauri-plugin-connector (Rust)
-
turbo-surf-mcp
turbo-surf MCP server — native Rust binary, stdio JSON-RPC (no Node)
-
mindreader
Selective durable graph memory MCP server for AI agents, backed by Neo4j
-
openlatch-client
OpenLatch runtime enforcement node — the capture-and-enforce adapter that evaluates every covered action against a coding agent's Autonomy Zone before it runs
-
mermaid-cli
Open-source AI pair programmer with agentic capabilities. Local-first with Ollama, native tool calling, and beautiful TUI.
-
openai-tools
Tools for OpenAI API
-
yana-rt
Yana AI Runtime — safety CLI for AI agents: scan, graph, vault, hunt, ci, map, fix, doctor
-
ava-agent
a personal AI assistant
-
areev-llm
Out-of-box LLM provider backends for Areev and Areev Loop: OpenAI-compatible, Anthropic, and Ollama adapters over a small blocking HTTP client, implementing areev_loop::LlmBackend…
-
processfork
CLI — the
pfbinary. Snapshot, branch, and merge live AI agent state. -
lemma
A pure, declarative language for business rules
-
ingot-syntax
Abstract syntax tree for the Ingot agent language
-
uninews
A universal news scraper for extracting content from various news blogs and news sites
-
frankensearch
Two-tier hybrid search for Rust: sub-millisecond initial results, quality-refined rankings in 150ms
-
runifold-model
Provider-neutral model protocol and streaming primitives for Runifold
-
modelsocket
protocol for efficiently integrating with LLMs
-
onde-mistralrs-code-exec
Python code execution for mistral.rs
-
deliver-cli
Streamlined MCP server for managing spec workflow (requirements, design, implementation)
-
run_code_rmcp
云函数服务,执行JS/TS/Python语言代码,脚本必须有约定的函数名称(handler/main),会调用约定的函数名称结果和日志返回
-
thryd
A light weight, embeded llm request router, with cache
-
elicitation
Conversational elicitation of strongly-typed Rust values via MCP
-
kaptein
CLI
-
git-alias
A fast git alias tool with dual-style command support
-
inowatch
File Watch Daemon — watches filesystem changes and emits NDJSON or serves MCP
-
aof-core
Core types, traits, and abstractions for AOF framework
-
alex
local LLM credential vault, multi-provider routing proxy, and trace capture daemon
-
decapod
Rust-built governance runtime for AI agents: repo-native state, enforced workflow, proof gates, safe coordination
-
vizier
A digital steward for Your Majesty
-
terraphim_agent_evolution
Agent evolution system for Terraphim AI
-
notarai
CLI validator for NotarAI spec files
-
agcp
Lightweight Rust proxy translating Anthropic Claude API to Google Cloud Code API
-
opty-tui-agent-bridge
Ratatui terminal interface for opty fleet reports. Renders fleet summary, agent roster, action plan, and selected issue list from a JSON payload supplied by the opty CLI launcher.
-
lean-ctx-protocol
Open wire contracts for lean-ctx: events, observations, assignments, evidence gaps
-
nika-lsp
Reserved name for the Nika project (the AI workflow language + reference engine). The real crate ships with the Diamond release — https://nika.sh
-
rto-remote
remote model tier's consent gate, payload allow-list and egress record (ADR-0019) — the policy half of Roteiro's one explicitly-consented egress path, with no transport of its own…
-
grain-agent-harness
Rust port of @earendil-works/pi-agent-core harness: session persistence, system-prompt assembly, truncation, custom message helpers
-
social_networks
my utils for social networks managing/monitoring
-
clausura-core
Core library for Clausura — a CI-native agent for deterministic pipeline gating
-
cersei-agent
Agent builder, agentic loop, streaming, and reporters for the Cersei SDK
-
youtrack-mcp-rs
MCP server for YouTrack — issues, articles, agile boards, and time tracking for AI coding agents
-
cargo-context-mcp
Model Context Protocol server exposing cargo-context as tools/resources/prompts
-
lc-core
Core abstractions for langchainrust — Runnable, BaseTool, BaseChatModel, etc
-
nsip
NSIP Search API client for nsipsearch.nsip.org/api
-
hs-predict
HS code prediction for chemical products — Akinator-style interactive classification with rule-based and LLM hybrid engine
-
loomweave-mcp
Loomweave MCP server: exposes the code-archaeology graph to consult-mode LLM agents
-
org-core
org-mode file operations with fuzzy search, outline extraction, ID-based lookups, and heading content retrieval
-
temote-mcp
A sandboxed local-machine MCP server with out-of-band approvals
-
typesec-rbac
RBAC policy engine for typesec — YAML → typed enforcement
-
brazen
A stateless, swiss-army-knife adapter for every LLM provider and protocol
-
ryu-mail
Standalone out-of-process mail sidecar for Ryu: self-host Agent Inboxes (receive/store/send) that Core proxies to on loopback. The tracer for apps-as-microservices.
-
api_ollama
Ollama local LLM runtime API client for HTTP communication
-
openai-api-rs
OpenAI API client library for Rust (unofficial)
-
anthropic-async
Anthropic API client for Rust with prompt caching support
-
zag-cli
A unified CLI for AI coding agents — Claude, Codex, Gemini, Copilot, and Ollama
-
auricle-pipeline
VAD gate, chunker, and transcript assembler for Auricle
-
cortex-mem-tools
High-level tools and utilities for Cortex Memory Integration
-
mcp2cli
Turn any Model Context Protocol server into a native command-line application — typed --flags from JSON Schema, multi-transport, daemon mode, and shell-friendly output formats
-
agentspec
Universal agent skill and sub-agent manager with TUI
-
agentsdk
A lean, OpenAI-compatible AI SDK for Rust
-
mcpd
Aggregates multiple MCP servers behind a single endpoint
-
mcpr-core
Core types, traits, protocol, and proxy engine for mcpr crates
-
compendious-machine
MCP server for text density optimization - maximize information per token
-
agentbus-stdio
stdio MCP bridge for agentbus, exposing the message bus as MCP tools
-
pctx_executor
TypeScript execution environment orchestration
-
gibberish-or-not
Figure out if text is gibberish or not
-
rust-agent-core
Core abstractions (Agent trait, Message, Session, Tool) for the rust-agent-framework ecosystem
-
va-ai-api-bridge
Protocol translation primitives for AI API request and response shapes
-
treehopper
Fuzzy-pick and manage git worktrees, with optional Claude Code session support
-
rolter-gateway
rolter data-plane gateway (openai/anthropic-compatible proxy)
-
sirno
gives project design a semantic intermediate representation
-
leann-core
LEANN is a revolutionary vector database that democratizes personal AI. Transform your laptop into a powerful RAG system that can index and search through millions of documents while…
-
seher-sdk
Seher SDK: agent resolution and codexbar-backed rate-limit checks
-
oxidelm
Bare-metal pure-Rust Transformer engine with reverse-mode autograd, BPE tokenization, and wgpu acceleration
-
leetgo-rs
Rust test utils for leetgo
-
adaq-news
Async financial news acquisition library (Cailianpress & Eastmoney) with a MacOS TTS CLI
-
episteme
Knowledge graph for software engineering — design patterns, refactorings, and laws for AI agents
-
omnigraph-engine
Runtime engine for the Omnigraph graph database
-
jerrycan-storage
Object-storage extension for the jerrycan framework: design-modeled buckets, local + S3-compatible blob stores, signed URLs, owner/tenant-scoped metadata. https://jerrycan.cc
-
blazen-events
Event traits and built-in event types for the Blazen workflow engine
-
context_weaver
A lorebook engine for LLM role-playing applications, built on weaver_lang
-
vibesurfer
A browser for LLMs, not humans
-
rma-common
Common types and utilities for Qryon
-
tt-inspect-rules-tier1
The 10 launch-day P0 Inspect rules (deterministic, Tier 1)
-
gbnf
working with GBNF
-
hrdr-tools
Agentic tool set for hrdr — read, write, edit, bash, grep, find, ls, todo, fetch, search, and MCP. Native OpenAI tool calls.
-
hexser
Zero-boilerplate hexagonal architecture with graph-based introspection
-
thulp
CLI for thulp - Execution context engineering platform
-
ariacompute-graph
Aria zero-copy compute graph
-
rkllm-rs
rkllm rust ffi binding
-
cap-rs-orchestrator
Fleet orchestration engine for CAP (CLI Agent Protocol) — declarative YAML-driven multi-agent coordination
-
zeph-memory
Semantic memory with SQLite and Qdrant for Zeph agent
-
acp-llm-adapter
ACP adapter server that exposes LLM providers (DeepSeek, GLM, Groq) as agents to ACP-capable editors
-
glep
Indexed grep + glob for AI agents
-
rustfs-mcp
RustFS MCP (Model Context Protocol) Server
-
tokf
Config-driven CLI tool that compresses command output before it reaches an LLM context
-
rememhq-mcp
MCP server for remem — exposes memory tools over stdio/HTTP
-
sema-otel
OpenTelemetry GenAI tracing facade for the Sema runtime
-
assay-monitor
Userspace monitor for Assay eBPF programs
-
ms-pdb-mcp
MCP server for analyzing PDB files using the ms-pdb library
-
zapmomo
An open-source, real-time desktop AI companion with voice, memory, and a customizable virtual character
-
clam-cli
Manage multiple Claude Code profiles
-
memoricai-core
Core domain types, wire DTOs, and provider ports for memoricai — a memory & context engine for AI agents
-
oneai-memory
OneAI memory system (short-term, long-term, context compression)
-
candle-mi
Mechanistic interpretability for language models in Rust, built on candle
-
autonomic-controller
Pure rule engine and projection reducer for the Autonomic homeostasis controller
-
cc-colony
Multi-agent orchestration for Claude Code with tmux and integrated MCP executor
-
gitz-cli
Craft intelligent Git commit messages with AI, enhancing your development workflow
-
adk-mcp-github
GitHub MCP Server — repositories, files, branches, issues, PRs, diffs, reviews, and releases for ADK-Rust Enterprise agents
-
apohara-agentguard
Deterministic, offline, no-model safety hook + local seccomp+Landlock sandbox + input firewall for Claude Code
-
kremis-core
Deterministic graph engine for grounded AI - every query returns fact, inference, or unknown. Never invents.
-
openmcpgdb
Interactive MCP server to control gdb. Fully featured and written in rust
-
agentkit-loop
Runtime-agnostic loop driver, interrupts, and model session traits for agentkit
-
llama-cpp-sys-2
Low Level Bindings to llama.cpp
-
pleiades-agent-providers
Pleiades AI provider implementations
-
hagent
A thin terminal agent client for Hadrian's OpenAI-compatible /v1/responses API
-
ariacompute-inference
Aria bundle load + session inference
-
consult-llm-monitor
Live TUI monitor for consult-llm runs
-
reflex-search
A local-first, structure-aware code search engine for AI agents
-
openai-rust2
An unofficial library for the OpenAI API (OpenAI-compatible endpoints, streaming chat, Responses API, images)
-
vtmate
audio ai conversation system
-
gemini-ai
based Gemini AI API, enabling access to advanced natural language processing and multimodal models
-
codexusage
Fast CLI reports for OpenAI Codex session usage and cost
-
etalon-mcp-server
ETALON Model Context Protocol (MCP) Server for AI native privacy analysis
-
copilot-api
OpenAI and Anthropic-compatible gateway for GitHub Copilot or Codex or third-party providers (Rust port)
-
agend-terminal
Agent Process Manager — orchestrate AI coding agents with PTY isolation, fleet management, and 32 MCP tools
-
gguf-utils
handling gguf files
-
llmy-harness-app
All-in-one LLM utilities
-
cnctd_ai
Multi-provider AI/LLM abstraction layer with MCP support (Anthropic, OpenAI, Gemini, OpenRouter)
-
greentic-llm
Provider-agnostic multi-LLM abstraction for the Greentic platform (rig-core backend, 24 providers)
-
lineark
CLI for Linear issue tracker — for humans and LLMs
-
talos-config
Configuration schema, validation, and environment substitution
-
morok-ir
Intermediate representation for the Morok ML compiler
-
claux
Terminal AI coding assistant with tool execution
-
ganit-mcp
MCP server exposing ganit formula evaluation as a tool for AI assistants
-
llm-bridge-core
Protocol transform library for LLM API translation between Anthropic and OpenAI
-
organism-runtime
Curated embedded runtime for Organism — registry, readiness, and pipeline wiring
-
smooai-smooth-operator
Reference core for smooth-operator: domain model + StorageAdapter seam over smooth-operator
-
api_openai
OpenAI's API for accessing large language models (LLMs)
-
opengate-models
Domain types, enums, and DTOs for OpenGate
-
mnm-telemetry
Opt-out telemetry event schemas, batching client, and three-mechanism opt-out
-
hyprwhspr-rs
Native speech-to-text voice dictation for Hyprland (Rust implementation)
-
scriptrs
Fast NVIDIA Parakeet TDT v2 and v3 speech-to-text in Rust, on Core ML (macOS) and native CUDA (Linux, Windows)
-
modular-agent-core
Modular Agent Core
-
unigateway-sdk
Thin facade crate for embedders using UniGateway core, protocol, and host libraries
-
codebones
An AST-aware code indexing and AI context extraction tool
-
gent-lang
A programming language for AI agents
-
spider-browser
Browser automation client for Spider's pre-warmed browser fleet with smart retry and browser switching
-
el-ffi
Host bindings: uniffi-bindgen-react-native (RN), flutter_rust_bridge-generated Dart, wasm-bindgen (web) — ADR-001/024
-
scouter-profile
Scouter profile logic
-
sinter-core
Typed code-graph model: nodes, edges, invariants
-
git-surgeon
Surgical git hunk control for AI agents
-
rigorix-engine
Template-driven DAG execution engine with bounded autonomy
-
dynamo-truthy
Canonical truthy/falsy boolean flag parsing for Dynamo
-
cardinal-harness
RENAMED: this crate is now
llmsort(lineage: cardinal-harness -> ratiometer -> llmsorting -> llmsort). Use llmsort >= 0.14 -
lutum-protocol
Core traits and request/response types for lutum
-
ffs-mcp
MCP server for ffs file finder — drop-in replacement for AI code assistant search tools
-
glyphtrail-cli
Glyphtrail: map codebases as semantic and historical graphs
-
harn-stdlib
Embedded Harn standard library source catalog
-
haskelujah-diagnostics
Diagnostics rendering and reporting utilities for the Haskelujah compiler
-
ras-events
Tokio broadcast event bus
-
agentzero-core
AgentZero — modular AI-agent runtime and tool framework
-
zero-commands
Command parser and dispatcher contracts for the ZERO operator CLI
-
mcp-methods
Reusable utility methods for MCP servers — pure-Rust library
-
anyllm_translate
Pure translation layer between Anthropic Messages API and OpenAI Chat Completions
-
onde
On-device inference engine for Apple silicon
-
tokenmaster
Core context-budget metering and decision engine for LLM applications: normalized token accounting, effective-budget gauges, turns-to-exhaustion prediction, and compaction/handoff decisions
-
glimpse
A blazingly fast tool for peeking at codebases
-
multi-llm
Unified multi-provider LLM client with support for OpenAI, Anthropic, Ollama, and LMStudio
-
openheim
A fast, multi-provider LLM agent runtime written in Rust
-
owly
A linter for open-knowledge format LLM wikis
-
tavily
A tiny tool for calling Tavily's REST API in the simplest way!
-
facto-mcp
MCP server library for facto: exposes package, forge, runtime, and lockfile tools over rmcp
-
dbx-tools-model-proxy
Multi-protocol Databricks model proxy
-
kanban-mcp
Model Context Protocol server for the kanban project management tool
-
wicked-estate-retrieve
Agent-facing retrieval tools with hybrid and semantic search for wicked-estate
-
panel-detect-cli
CLI for the panel-detect runtime
-
siumai-spec
Provider-agnostic specs and data types for siumai
-
sapphire-agent-cli
The command line of sapphire-agent's two editions, the headless one and the desktop app: its commands, the framework's and its own, which run the agent or talk to it
-
nexus-memory-lephase
LePhase integration wrapper for Nexus Memory System
-
llm_serve
将 OpenAI 兼容接口的模型绑定 Tool,MCP,Skills 后发布为新的 OpenAI 兼容 API 服务
-
nitpicker-agent
Reusable agentic loop (file/git tools + parallel subagents) extracted from nitpicker
-
mem7
LLM-powered long-term memory engine — Rust core with Python bindings
-
llm_api_access
A package to query popular LLMs
-
coraline
semantic code knowledge graph for faster AI-assisted development
-
synaptic-middleware
Middleware system for Synaptic agents: AgentMiddleware trait and built-in middlewares
-
squick-format
LLM-targeted output formatters for Squick: compact markdown and JSON tailored for AI agents
-
RavenClaws
Lightweight, secure Rust agent framework with multi-provider LLM support
-
bamboo-infrastructure
Process management, logging, and metrics for the Bamboo agent framework
-
orchestrator-collab
Agent collaboration types — shared context, DAG, and artifacts
-
liter-llm-cli
CLI for liter-llm — start an OpenAI-compatible proxy server or MCP tool server
-
mcpsec
MCP Security Benchmark Framework — vendor-neutral security evaluation for MCP gateways
-
ecl-pipeline
Pipeline execution engine for ECL pipeline runner
-
iab-specs
Unofficial Rust implementation of IAB specifications: OpenRTB 2.5/2.6/3.0, AdCOM, Ads.txt, Sellers.json, ARTB, Agentic Direct, and more. Umbrella crate with feature-gated sub-crate re-exports.
-
tldr-mcp
MCP server for TLDR code analysis tool
-
pitboss
CLI that orchestrates coding agents (Claude Code and others) through a phased implementation plan, with automatic test/commit loops and a TUI dashboard
-
ggen-config
Configuration parser and validator for ggen.toml files
-
terraphim_agent_supervisor
OTP-inspired supervision trees for fault-tolerant AI agent management
-
gvrn
DEPRECATED — renamed to
ductus. Installductusinstead; this crate is no longer updated. Final release of the govern-era runtime. -
oxo-flow-web
Web interface for the oxo-flow bioinformatics pipeline engine
-
gossiphs
lib for general code file relationship analysis. Based on tree-sitter and git analysis.
-
highflame
Rust SDK for the Highflame AI guardrails service
-
gitbuddy
An AI-driven tool designed to simplify your Git commit process
-
data-modelling-core
Core SDK library for model operations across platforms
-
mcp-student-records
Student Records MCP Server — a Student Information System (students, courses, enrollments, grades & rubrics, attendance, support, guardians & FERPA-gated communications, transcripts…
-
perspt-agent
SRBN Orchestrator and Agent logic for Perspt
-
codewhale-mcp
Bounded MCP guidance normalization for Codewhale
-
pensieve-brain
Brain repos: deterministic memory→Obsidian-vault export, git-binary wrapper, smart-HTTP pkt-line helpers, and push-ingest planning — the pure library behind pensieve's git-clonable brains
-
mcp-confluence
MCP server for Confluence integration - create, update, search, and manage Confluence pages
-
ollama-api-rs
An async Rust SDK for the Ollama API with OpenAI compatibility
-
a3s-ahp
Agent Harness Protocol v2.4 — Universal, transport-agnostic protocol for supervising autonomous AI agents
-
tokn-endpoint-core
Shared endpoint primitives and traits used by tokn endpoint crates
-
ds-api
client library for the DeepSeek API with support for chat completions, streaming, and tools
-
heliosdb-codekb-mcp
Local MCP server for code+docs knowledge bases, embedding HeliosDB-Nano as a library
-
encoderfile
Distribute and run transformer encoders with a single file
-
aiway-protocol
The aiway protocol lib
-
flyllm
unifying LLM backends as an abstraction layer with load balancing
-
mcpeek
A TUI MCP inspector
-
codewandler-flux-credentials
flux provider credentials: token store, CLI-credential import, OAuth refresh + PKCE login, TokenSource impl
-
sqry-mcp
MCP server for sqry semantic code search
-
ts-pack-cli
CLI tool for tree-sitter language pack - manage parser downloads and run code intelligence
-
lunaris-embed-native
Native candle-rs ModernBERT FP16 embedder for IBM granite-embedding-311m-multilingual-r2
-
zeptoclaw
Ultra-lightweight personal AI assistant
-
writestead
LLM Wiki
-
rtk-context-mcp
RTK — Model Context Protocol (MCP) server for local context tools
-
harness-rs-models
LLM adapters for harness-rs — three protocol families via ApiKind::build(base_url, model, key): OpenAI-compatible, Anthropic Messages, Gemini native (Google Search grounding on by default)…
-
genify
Turn one file into a complete project
-
ai-memory
AI-agnostic persistent memory system — MCP server, HTTP API, and CLI for any AI platform
-
spider-rules
Spider/crawler identification rules
-
mcp-lightsail-rust-aws
Hello World streaming http MCP
-
forgejo-mcp-rs
Model Context Protocol server for Forgejo, Codeberg and Gitea, in Rust
-
lspz
AI-friendly LSP compression proxy
-
dnsync
DNS Sync and Control with MCP
-
starweaver-tools
Function tools, tool registries, and execution primitives for Starweaver
-
turul-mcp-builders
Runtime builders for MCP (Model Context Protocol) components
-
aiprog
AI program runtime and API set (from Function Calling to Program Running)
-
a3s-sentry
Tiered (L1 rules / L2 LLM / L3 agent) runtime security control for AI agents, built on a3s-observer
-
evolve-adapters
Adapter trait + per-tool integrations (Claude Code, Cursor, Aider)
-
copilot-sdk-supercharged
Rust SDK for programmatic control of GitHub Copilot CLI via JSON-RPC 2.0
-
mermaid-model
Model-layer core for Mermaid: provider adapters, wire types, capability catalog, and the shared utilities they need
-
cowork
CLI tool for managing Claude Code skills - install, generate, and organize skills across AI coding agents
-
kq-llm
LLM provider abstraction for kq: Ollama, OpenAI and Anthropic backends
-
pyannote-rs
Speaker diarization using pyannote in Rust
-
nanny-bridge
Internal runtime component of nanny, the authorization and audit layer for AI agents
-
ollama-kernel
Ollama Jupyter Kernel
-
uira-core
Shared types, events, protocol definitions, and configuration loading for Uira
-
aidaemon
A personal AI agent that runs as a background daemon, accessible via Telegram, Slack, or Discord, with tool use, MCP integration, and persistent memory
-
aquaregia
A lightweight agent SDK for Rust: build tool-using LLM agents that run on any provider
-
cf-gears-rustls-fips-shim
Workspace-feature-unification shim: pulls rustls + hyper-rustls with the
fipsfeature on every target where the AWS-LC FIPS backend is the intended TLS provider (i.e. everywhere except macOS… -
astrid-config
Unified configuration system for Astrid
-
oxibrain-mcp
MCP server tools for oxibrain — fifteen-tool cap
-
pakx-cli
Universal package manager for AI agent context (binary crate; installs the
pakxcommand) -
slimshady
Strip environment variable values for safe LLM sharing
-
bote
MCP core service — JSON-RPC 2.0 protocol, tool registry, audit integration, and TypeScript bridge
-
cf-gears-llm-gateway-sdk
SDK for the LLM Gateway: Open Responses domain models
-
quorum-rs
Rust SDK and CLI for multi-agent deliberation systems — ships the
quorumbinary (run / status / trace / tui / init) plus the underlying agent, LLM, tool, prompt, and worker library -
vibe-tune
Convert codebases into high-quality LLM instruction-tuning datasets. AST-based extraction, domain filtering, quality scoring, multi-modal outputs.
-
agentforge-runner
Agent execution engine: parallel workers, trace capture, retry logic (F-03)
-
iron-providers
Semantic provider boundary for protocol-oriented LLM providers
-
blz-mcp
MCP server for blz documentation search
-
crabmate
Rust AI agent: OpenAI-compatible chat/completions, function calling, HTTP serve, ops CLI
-
repocat
concatenate all code and text files in a github repo for LLM inference contexts
-
liboxen
Oxen is a fast data version control system, built with machine learning training data in mind. Designed to handle terabytes of data with ease, using a workflow similar to git. Version…
-
toktrie_hf_tokenizers
HuggingFace tokenizers library support for toktrie and llguidance
-
agx-mcp
MCP server exposing agx's session introspection tools to AI agents (Claude Code, Cline, Gemini CLI)
-
streamdown-parser
Streaming markdown parser for streamdown
-
jumble
An MCP server that provides queryable, on-demand project context to LLMs
-
fathom-mcp
CLI + MCP server wrapping the OpenAI Deep Research API
-
datahugger
fetching data and metadata from DOI or URL
-
signalwire-sdk
SignalWire AI Agents SDK for Rust — build AI voice agents, control live calls over WebSocket, and manage resources over REST
-
siumai-protocol-openai
OpenAI Chat Completions, Responses, and Realtime wire codecs for siumai
-
longline
System-installed safety hook for Claude Code and Codex CLI
-
twofold
One document, two views. Markdown share service for humans and agents.
-
edgee
Rust SDK for the Edgee AI Gateway
-
ferro-mcp
MCP (Model Context Protocol) server for AI-assisted Ferro Framework development
-
loong-bridge-runtime
Internal support crate for Loong: reusable managed bridge execution primitives
-
dynamo-async-openai
Fork of async-openai customized for Dynamo
-
meerkat-runtime
v9 runtime control-plane for Meerkat agent lifecycle
-
pf-core
ProcessFork core: content-addressed store, .pfimg format, atomic snapshot orchestration
-
nemo-relay-pii-redaction
First-party deterministic and model-backed PII redaction plugin surfaces for NeMo Relay
-
polyforge-mcp
MCP server exposing the PolyForge evidence ledger and completion gate over stdio (rmcp)
-
rho-hashline
Line-level hashing for file editing anchors
-
open-multi-agent-rs
A multi-agent orchestration framework for Rust — streaming LLM calls, tool use, task pipelines, retry, and observability
-
swiftide-query
Composable LLM agents, typed task graphs, and streaming RAG pipelines in Rust
-
agents-core
Core traits, data models, and prompt primitives for building deep agents
-
atman-proto
atman JSON-RPC 2.0 protocol types for the headless daemon
-
g3-core
Core engine for G3/GB AI coding agent
-
langchain-rust
LangChain for Rust, the easiest way to write LLM-based programs in Rust
-
orra
Context-aware agent session management for any application
-
lattice-embed
SIMD-accelerated vector operations and embedding generation
-
vv-llm
surface for vv-llm
-
egui-mcp-server
MCP server for egui UI automation
-
shell-agent
AI agent that only knows how to run shell commands
-
rtk-registry
Command rewriting engine from the RTK (Rust Token Killer) project
-
nav-cli
A semantic directory bookmarking tool built for both humans and AI Agents
-
bitbucket-dc-mcp
Bitbucket Data Center MCP server, generated by mcpify
-
tandem-types
Shared Tandem data types
-
onwards
A flexible LLM proxy library
-
pareto-rs
Pareto analysis helpers for tokenledger
-
qtcloud-data-cli
量潮数据云 CLI
-
pizen-lynx
Lynx: deterministic repository intelligence for AI-native tooling. Ships the
lxCLI and thelx mcpstdio MCP server. -
unbrowser
Web access for LLM agents. One static binary. No Chrome.
-
slm_inference
Backend-agnostic trait layer for running Small Language Model (SLM) inference in Rust
-
lunaris-rerank-native
Native candle-rs XLM-RoBERTa FP16 cross-encoder for BAAI/bge-reranker-v2-m3
-
edda
Decision memory for coding agents — CLI and TUI
-
crw-crawl
Async BFS web crawler with rate limiting and robots.txt support for CRW
-
paraoxidizer-quant
Quantization kernels, group-wise INT4/INT8, outlier tables, and SIMD acceleration
-
bitrouter
LLM API router. CLI + assembly library.
-
raycast-mcp-server
MCP server for Raycast automation on macOS -- 9 tools for search, commands, clipboard, shortcuts, window control, system functions, auth, extensions, workflows
-
llmkit
Production-grade LLM client - 100+ providers, 11,000+ models. Pure Rust.
-
rig-bedrock
AWS Bedrock model provider for Rig integration
-
awaken-stores
Storage backends (memory, file, PostgreSQL, SQLite mailbox) for Awaken agent state
-
prigen-cli
PriGen es una herramienta para generar proyectos de manera rápida y eficiente siguiendo los lineamientos de arquitectura
-
better-ctx
Context Intelligence Engine with CCP. 34 MCP tools, 8 read modes, 90+ compression patterns, cross-session memory (CCP), persistent AI knowledge with temporal facts + contradiction detection…
-
aethershell
The world's first multi-agent shell with typed functional pipelines and multi-modal AI
-
ironflow-mcp
MCP server for Ironflow workflow orchestration
-
symposium
AI the Rust Way
-
ares-cli
CLI entry point for Ares AI scraper
-
langgraph
building stateful, multi-actor applications with LLMs (Full suite)
-
toktier-routing-core
Allocation-free per-request routing and certified BPE boundary selection for TokTier
-
every-other-token
Watch an LLM answer token by token: a CLI, terminal UI and web UI that intercepts the OpenAI, Anthropic, Gemini, OpenRouter or local (Ollama, OpenAI-compatible) stream, shows per-token…
-
bouncy-cli
Tiny Rust headless browser. Scrape pages or drive multi-step browse sessions (click / fill / submit / read) from the CLI. Same workspace also ships an MCP server (bouncy-mcp) for LLM-driven autonomous browsing.
-
rswarm
Swarm framework
-
llmux
Hook-driven LLM model multiplexer with pluggable switch policy
-
plasmate
Agent-native headless browser engine with Semantic Object Model
-
peon-runtime
A runtime-agnostic agent execution engine with zero-trust context injection, multimodal messaging, and pluggable LLM providers
-
langgraph-checkpoint-rs
Core checkpointing traits and interfaces for LangGraph applications in Rust
-
hrdr-editor
FSM-agnostic text-editor seam for hrdr, embedding the hjkl engine behind an EditorEngine trait
-
rullama-provider
LLM provider implementations (Anthropic, OpenAI Chat + Responses, Google Gemini, Ollama, Bedrock, Vertex AI, local llama.cpp) for the rullama agent framework. Speech (TTS/STT) providers live in
rullama-provider-speech. -
truenas-master-mcp
Official MCP server for TrueNAS API access
-
piko-types
Core types and data structures for PikoClaw
-
loomweave-federation
Loomweave federation: shared HTTP types and config for Filigree/Wardline sibling exchange
-
langrust
Unified API for LLM providers
-
rapidgate
High-performance unified LLM API gateway written in Rust
-
redshank-cli
CLI entry point for the Redshank investigation agent
-
wdpkr
Codebase search via vector retrieval over LLM-generated summaries, exposed as a CLI for AI coding agents
-
mcp-router
An MCP Router for the masses. Inspired by the myriad of difficult to use, limited use cased libraries currently available
-
ai-agent
Idiomatic agent sdk inspired by the claude code source leak
-
towl
A fast CLI tool to scan codebases for TODO comments and output them in multiple formats
-
punch-kernel
The Ring — central kernel and orchestrator for the Punch Agent Combat System
-
edgecrab-types
Shared types for the EdgeCrab agent: messages, tool schemas, errors, config types
-
areev-mcp
Model Context Protocol (MCP) stdio server for Areev
-
agentty
ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
wicked-estate-memory-api
The memory/knowledge interface seam: traits + DTOs a host (estate binary/MCP) calls so any memory implementation (e.g. wicked-memory) can plug in. Depends on NOTHING heavy; estate-core does NOT depend on this.
-
agentflare
Optimize AI CLI agents for cost and performance
-
liter-llm-bindings-core
Shared utilities for liter-llm language bindings — case conversion, config parsing, error formatting, runtime management
-
dkp
dkp — Domain Knowledge Pack management CLI
-
atuin-ai
AI integration for Atuin CLI
-
opensymphony
OpenAI Symphony orchestration design
-
naptrace
Variant analysis, open-sourced. Feed a CVE patch, find its structural twins.
-
logprobe
Detect normalization errors, entropy bias, and truncation artifacts in LLM logprob data
-
turul-mcp-protocol-2025-11-25
Model Context Protocol (MCP) specification implementation - 2025-11-25 version
-
googleapis-tonic-google-ai-generativelanguage-v1beta
A Google APIs client library generated by tonic-build
-
caliban-agent-core
Agent loop, tool dispatch, cancellation, retry, compaction, and hooks for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
red-queen-cli
Command-line interface for Red Queen
-
localgpt-world-types
Unified world data model for LocalGPT Gen — serde-only, no Bevy
-
dora-openai-proxy-server
doragoal is to be a low latency, composable, and distributed data flow -
pw-core
Shared types for PromptWallet - OpenAI-compatible chat API types, artifacts, and extensions
-
codepack
A *lightning-fast* tool that transforms a directory into a single, LLM-friendly text file, perfect for easy processing and analysis
-
xet-runtime
Async runtime, configuration, logging, and utility infrastructure for the Hugging Face Xet client tools
-
autonomic-core
Core types, traits, and errors for the Autonomic homeostasis controller
-
memra-tokenizer
GGUF-native BPE/SPM tokenizer with chat-template support for the memra inference engine
-
llm_runtime
Abstractions and primitives for building agents and runtimes on top of llm_adapter
-
polyc-llm-vertex
LlmProvider backend over the GCP AI platform REST API (WIF/ADC auth)
-
cloakpipe-core
Regex PII detection, pseudonymization and rehydration engine for CloakPipe (NER behind the
nerfeature) -
rig-model-catalog
Provider-agnostic model metadata (context window, capabilities, quantization) and probe traits for Rig agents
-
modelexpress-client
Client library for Model Express gRPC server
-
semantex-mcp
MCP server for semantex semantic code search
-
tt-core
Axum HTTP server, routing engine, middleware, and provider registry
-
vigil-pulse
Metacognitive monitoring for AI entities — pipeline enforcement, reflection quality, outcome tracking
-
houston-skills
Hermes-style self-improving skills for AI agents
-
lethe-claude-code
Claude Code adapter for the lethe memory store. Helpers used only by the plugin under
plugins/claude-code/(transcript parsing, etc.). Not required to use lethe itself. -
hrdr-tui
hrdr terminal UI — agent transcript + vim-keybound input pane (hjkl), live streaming
-
ironclaw_common
Shared types, paths, and platform helpers used across the IronClaw workspace
-
mika-common
Shared library for Mika AI assistant
-
latexsnipper-api-types
Public API types for LaTeXSnipper Core
-
dumbo-rs
Turn any codebase into LLM-ready context : supports monorepos, multi-project runs, and git diffs
-
ai-adapter
AI API adapter - bidirectional proxy translating between OpenAI Responses, Chat Completions, and Anthropic Messages APIs
-
spider-cloud-cli
The Spider Cloud CLI for web crawling and scraping
-
you
Translate your natural language into executable command(s)
-
mcp-langbase-reasoning
MCP server providing structured reasoning via Langbase Pipes - linear, tree, divergent, Graph-of-Thoughts, and decision framework modes
-
Zhu-li
that monitors for new video or audio files, sends them for transcription, then summaries the docs and adds them to an obsidan vault
-
rwkv-tokenizer
A fast RWKV Tokenizer
-
rust-hf-downloader
TUI for searching and downloading HuggingFace models
-
tokn-catalogue
Model and provider catalogue loading and mapping utilities for tokn
-
cerememory-engine
Orchestrator that assembles all Cerememory stores and engines
-
nucl-parquet-mcp
MCP server for nuclear data — wraps local Parquet via nucl-parquet crate
-
lellm-agent
Agent Runtime for LeLLM — ToolUseLoop, Executor, Fallback
-
evo-common
Shared types, protocols, and utilities for the Evo self-evolution agent system
-
shore-protocol
Shore Wire Protocol (SWP) message types and serialization for the Silvershore chat daemon
-
golem-ai-llm-openai
working with OpenAI APIs on Golem Cloud
-
keel-cli
CLI entry point and command routing for keel
-
drut-mcp
Drut MCP server: read-only diagnose/format/query_structure/lookup_keyword tools over voyager-core
-
provekv
— two-tier, receipted, content-addressed KV-cache pool. fib_k4_n32 cold tier + turbo_8bit hot tier. The pool is the system; the codecs are the primitives.
-
vllm-doctor
Diagnostic tool for vLLM inference servers
-
llmenv-config
Configuration handling for llmenv
-
ras-crawl
Reusable Rust crawling, extraction, and optional GUI building blocks
-
adaclaw
Lightweight, secure, multi-channel Rust AI Agent Runtime
-
agent-shield
Security scanner for AI agent extensions - offline-first, multi-framework, SARIF output
-
data-gov-mcp-server
MCP Server for AI integration with Data.Gov
-
spinr
HTTP performance & debugging tool
-
odir
Ollama Downloader in Rust (ODIR), pronounced oh dear, is a command-line tool written in Rust for downloading models from Ollama
-
agentbus-cli
Command-line client for agentbus, an MCP-native message bus for AI agents
-
zeroclaw-tool-call-parser
Tool call parsing for LLM responses — handles JSON, XML, GLM, MiniMax, Perl-style, and more
-
servo-fetch-cli
A browser engine in a binary — fetch, render, and extract web content as Markdown, JSON, or screenshots. Powered by Servo.
-
colossus
powerful tool for AI-driven text processing and analysis
-
synapse-cli
CLI for Synapse - Configuration Language for Memory Systems
-
talos-core
Foundation types, core traits, and error definitions
-
nutrition_rs
A CLI for flexible plaintext nutrition tracking
-
oris-mutation-evaluator
Mutation quality evaluator with static analysis and LLM critic
-
cortex-mem-mcp
MCP server for Cortex-Mem memory management
-
cvkg-gallery
Cyber Viking Kvasir Graph (CVKG) - High-fidelity agentic UI framework
-
zellig
Auto Translation CLI tool with local and AI support
-
klava
CLI tool for using code agents with any provider. Use Claude Code with your OpenAI-compatible provider. Make code agents more secure by filtering out leaked secret keys and cryptographic keys from your filesystem.
-
fluers-runtime
The Fluers agent harness: agent definition, sessions, skills, sandbox, events
-
llm-sdk-rs
that enables the development of applications that can interact with different language models through a unified interface
-
dkp-gen-core
LLM-driven DKP generation pipeline library
-
dar-cap-runner
Runner capability contract and shared runner environment types
-
papercuts
A tiny CLI for AI agents to log the papercuts they hit during work
-
codewandler-flux-markdown
flux's own markdown engine: frontmatter parsing plus an AST-based CommonMark-subset parser, writer, and (feature-gated) ratatui/terminal renderers
-
oneai-parser
OneAI 3-layer output parsing defense (constrained decoding, fuzzy repair, fallback)
-
auricle-vision
On-demand active-window capture (Windows.Graphics.Capture) + local OCR (Windows.Media.Ocr) for Auricle
-
lux-core
A terminal-native music CLI replacing lx-music-desktop, powered by Agentic intelligence
-
gephyr
headless local AI relay/proxy API handling OpenAI, Claude, and Gemini-compatible APIs
-
swarms-rs
The Bleeding-Edge Production-Ready Multi-Agent Orchestration Framework in Rust
-
just-openai-compat
OpenAI-compatible API client and wire-level types
-
dk-runner
dkod verification runner — CI/CD pipeline execution
-
furnace-mcp
FURNACE - AI-powered code generation and analysis MCP server with knowledge graph and tiered search
-
ytx
A command line utility that generates articles from youtube videos
-
scute-test-utils
Test utilities for Scute
-
rustcodegraph
Local-first code intelligence graph and CLI for AI agents
-
langchain-ai-rust
Build LLM applications in Rust with type safety: chains, agents, RAG, LangGraph, embeddings, vector stores, and 20+ document loaders. A LangChain port supporting OpenAI, Claude, Gemini…
-
otelite-client
Typed HTTP client for the Otelite REST API
-
deepsearch
A CLI for AI-powered research assistance, using local LLMs to decompose questions, search the web, and synthesize answers
-
orca-ai
AI operations assistant with conversational alerts and LLM-powered diagnostics
-
modelcontextprotocol-server
Server implementation for the Model Context Protocol (MCP)
-
wordchipper
HPC Rust LLM Tokenizer Library
-
unrager
Takes the rage out of your x.com timeline with a local LLM: browser extension, terminal client and API server
-
episodic
Reusable Observational Memory core models and pure transforms
-
stock-analyzer
Stock analysis engine — CLI & MCP server for market guidance, stock selection, and analysis reports
-
harn-flock
Advisory file locks with deadlines for the Harn programming language
-
kora-mcp
Model Context Protocol client for the Kora runtime
-
kvlar-cli
CLI tool for Kvlar — manage policies, query logs, run local security proxy
-
FerriteChatter
ChatGPT CLI
-
stakpak-agent-core
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
problems99
AI-native CLI for issue/PR context retrieval across Jira, GitLab, GitHub, and Bitbucket, with structured output for humans and AI agents
-
punch-runtime
Agent execution engine for the Punch Agent Combat System
-
crow-streamdown-parser
Streaming markdown parser for streamdown (crow fork: upstream fixes not yet released upstream)
-
ras-sandbox
Sandboxed code execution
-
xchecker-redaction
Secret detection and redaction for xchecker
-
kreuzcrawl-bindings-common
Shared adapter layer for kreuzcrawl language bindings
-
api_claude
Claude API for accessing Anthropic's large language models (LLMs)
-
synaptic-tools
Tool registry and execution: ToolRegistry, SerialToolExecutor, ParallelToolExecutor
-
scix-client
SciX (NASA ADS) literature search as an MCP server, CLI, Rust library, and Python module
-
kimetsu-core
Shared core types (config, events, ids, paths, memory kinds) for the kimetsu agent runtime + brain
-
rho-agent
AI coding agent with file tools
-
cartomancer
PR review tool with blast radius awareness — opengrep + cartog + LLM
-
nomoreide-remote-protocol
The frozen wire protocol between a local NoMoreIDE daemon and the hosted remote-control relay
-
agentkit-provider-openai
OpenAI provider adapter for agentkit
-
llm-relay
Shared types, format conversion, and HTTP client for Anthropic and OpenAI LLM APIs
-
mcp-marketing
Marketing Orchestration MCP Server — multi-channel campaigns, audiences, content, ads, performance, budget governance for AI agents
-
smos
SMOS adapter layer — concrete implementations of port traits
-
reson-agentic
Agents are just functions - production-grade LLM agent framework
-
tldrs
README.md generator powered by LLMs and codebase analysis
-
organism-intent
Intent contracts for Organism — packets, admission, decomposition, resolution
-
komando
A command line utility to better organize and keep track of your commands
-
swiss-weather-mcp
MCP server for Swiss weather, pollen, radiation, UV, and air-quality data
-
terminal-aichat
A cli for AI/LLM chat in terminal. Extremely simple and easy to use. Using OpenAI-compatible
/v1/chat/completionAPI -
url-preview
High-performance URL preview generator for messaging and social media applications
-
axonml-distributed
Distributed training utilities for the Axonml ML framework
-
skg-turn
Shared toolkit for building operators — Provider trait, types, context strategy
-
sapphire-agent-server
sapphire-agent’s app server: the agent itself, with its channels, tools, memory and IPC, which
sapphire-agent serveruns -
stoa-cli
Stoa command-line interface
-
devboy-executor
Tool execution engine for devboy-tools — provider factory, enrichment pipeline, typed output. Wires every devboy provider into a uniform tool surface.
-
lean-host-mcp
MCP server hosting Lean 4 via a supervised lean-rs-worker child
-
fetchium-cli
Fetchium CLI — the universal retrieval layer for the internet
-
agentic-aegis-cli
CLI for AgenticAegis streaming validation
-
silicon-omni
Rust client for the silicon omni daemon
-
agentcontainer
A standard way to declare and run agent containers for your projects
-
lix-storage-rocksdb
Embeddable version control for apps and AI agents
-
dk-engine
dkod code analysis engine — semantic parsing, indexing, and search
-
car-engine
Core runtime engine for Common Agent Runtime
-
oxvessel
Developer release announcement tool
-
mnm-embedding
VoyageAI embedding and reranking client
-
kobito
Autonomous coding agent orchestrator — works while you sleep, like the elves in the shoemaker's tale
-
message-optimizer-bin
Standalone CLI for GPT-5.4 message optimization
-
merlion-core
Core message, tool, and agent-loop types for Merlion Agent
-
shack-gateway
Progressive-discovery, security-hardened MCP gateway and aggregator
-
hanzo-sandbox
OS-level sandbox for subprocesses spawned by hanzo (code execution, tools)
-
rs_openai
The OpenAI Rust library provides convenient access to the OpenAI API from Rust applications
-
jerrycan-observe
Observability extension for the jerrycan framework: request IDs, JSON logs, /healthz, Prometheus /metrics. https://jerrycan.cc
-
agentic-planning
Persistent intention infrastructure for AI agents
-
gollum-kb
Knowledge Base interface for the Gollum platform
-
daimon
A Rust-native AI agent framework
-
llmy
All-in-one LLM utilities
-
schemalint
Static analysis tool for JSON Schema compatibility with LLM structured-output providers
-
open-kioku-format
Prompt-oriented renderers for Open Kioku reports
-
git-hunk
Non-interactive hunk staging for AI agents
-
shift-proxy
Native Rust HTTP proxy for SHIFT — intercepts AI API requests, optimizes image payloads, and forwards to upstream providers
-
mcp-server-wezterm
MCP server for WezTerm terminal
-
readseek
structural digest for source files, images, and documents
-
self-llm
Unified chat API for multiple LLM providers
-
vibepod
Safely run AI coding agents in Docker containers
-
hbp100
privacy preserving system for llm
-
anvilforge-boost
Boost — Anvilforge's AI-agent toolkit. Embedded MCP server with framework introspection tools (routes, models, components, schema, queries, logs).
-
healthpoint-export
Export helpers and provenance manifests for Healthpoint records
-
astrid-mcp
MCP client with server lifecycle management for Astrid
-
mcp-rss
MCP server that provides RSS tooling
-
tokn-gateway-cli
CLI and local gateway runtime for serving OpenAI-compatible APIs via tokn
-
rullama
— the agent framework; build any AI application in Rust
-
ingot-language-service
Editor-facing analysis for Ingot: the diagnostics, hovers and symbols an editor shows, from the same compiler the CLI runs
-
pai_openai_adapter
OpenAI-compatible API adapter for PAI-Kernel · pre-request gate evaluation + post-response witness chain entry · reference middleware
-
gptengage
Multi-AI CLI orchestrator for Claude Code, Codex, and Gemini
-
groq-api-rust
ability to interact with the Groq API
-
grain-llm-models
Standardized LLM model registry for grain-agent. Embeds a vendored snapshot of model metadata (context window, capabilities, pricing, thinking profile); the
fetchfeature pulls live data from models.dev. -
shuru-store
NBD-backed storage layer for shuru microVMs
-
kokoro-en
A lightweight, offline Rust inference library for Kokoro TTS - an 82M-parameter open-weights text-to-speech model
-
ollamaMQ
High-performance Ollama proxy with per-user fair-share queuing, round-robin scheduling, and a real-time TUI dashboard
-
edgecrab-tools
Tool registry, ToolHandler trait, and 50+ tool implementations
-
tokitai-context
Git 风格平行上下文管理系统，专为 AI 智能体设计（可选集成 FileKV 高性能存储引擎）
-
agentrs
Facade crate and prelude for agentrs
-
granite-cli
CLI for discovering, configuring, and launching AI workflows powered by IBM Granite models
-
xa
Execute Anything via LLM - A CLI tool for arbitrary text processing using LLMs
-
zag-orch
Orchestration library for zag — multi-session coordination for AI coding agents
-
nu-mcp
Modules to run a model context protocol (MCP) server that provides Nushell as a tool
-
bito
Quality gate tooling for building-in-the-open artifacts
-
nomograph-sysml
CLI-native knowledge graph toolkit for SysML v2
-
leakferret-mcp
MCP (Model Context Protocol) server for leakferret — exposes scan, classify, propose_rewrite, verify, baseline_diff over JSON-RPC stdio
-
archidoc-types
Shared types for the archidoc toolchain
-
gemini_proxy
A high-performance, stateless OpenAI-compatible API gateway proxy for gemini-cli
-
defect-obs
Observability layer for defect: session events and Langfuse trace ingestion
-
oxicode-ai
Unified LLM API — multi-provider streaming interface for AI coding assistants
-
codex-budget-guard
Budget enforcement for OpenAI Codex CLI — daily/weekly/monthly token limits, phase detection, auto-throttle, and Serde auditing snapshots
-
kernex-agent
CLI dev assistant powered by Kernex runtime
-
rig-milvus
Milvus vector store implementation for the rig framework
-
rememhq-core
Core library for remem — reasoning memory layer for AI agents
-
qwed
Rust SDK for QWED Verification Protocol
-
codex-empirica-translator
ecodex's protocol translator — bridges OpenAI Chat Completions providers (DeepSeek, Qwen, Kimi, Ollama, etc.) to codex's Responses API surface
-
svipall-dashboard
Human dashboard for manual captcha solving
-
devboy-confluence
Confluence self-hosted provider for devboy-tools — KnowledgeBaseProvider/Provider implementation for Atlassian Confluence
-
rllm-core
Common configs, errors, IDs, request/response types, and sequence state for rLLM
-
sema-fmt
Code formatter for Sema source files
-
ryu-tuition
Tuition for Ryu: a single-learner tutor that owns its whole spine out-of-process — subjects and parsed source documents, a prerequisite DAG of skills each carrying a Bayesian Knowledge Tracing posterior…
-
objectiveai-api
ObjectiveAI API Server
-
hallouminate-config
Config layer for hallouminate
-
scud-cli
Fast, simple task master for AI-driven development
-
oxicode-sdk
oxicode AI agent SDK — build isolated, multi-agent AI systems
-
vibe_coding_tracker
Vibe Coding Tracker - AI coding assistant telemetry/usage parser, aggregate local session data into CodeAnalysis results
-
juglans
Compiler and runtime for Juglans Workflow Language
-
ghost_git_writer
write a git commit message, README or Diff Summary by LLM services
-
thruline-mcp
MCP stdio server exposing thruline state as tools
-
bitpolar-cli
Command-line tool for BitPolar vector quantization
-
k4000
AI-assisted feature development and code review MCP servers
-
ncp-runtime
NCP reference runtime — composable, auditable WASM agent graphs
-
llm-mux
Multiplexer for LLMs - route prompts to multiple backends
-
cargo-context-core
Core engine for cargo-context: assembles, budgets, and scrubs Rust project context packs for LLMs
-
cognate-cli
CLI tool for interacting with LLM providers via Cognate
-
embedrs
Unified embedding — cloud APIs (OpenAI, Cohere, Gemini, Voyage, Jina, Mistral) + local inference, one interface
-
ai-agents-llm
LLM providers for AI Agents framework
-
serdes-ai-models
Model trait and provider implementations for serdes-ai
-
clawft-types
Core types for the clawft AI assistant framework
-
kastellan-core
Agent core: scheduler, memory orchestration, policy gate, LLM router, IPC, audit log
-
zeph-mcp
MCP client with multi-server lifecycle and Qdrant tool registry for Zeph
-
rullama-reasoning
Layer 3 reasoning — plan/output parsers + provider-agnostic local-inference scorers (router, complexity, relevance, retrieval, strategy, summariser, validator, entity enhancer) for the rullama agent framework
-
org-mcp-server
Model Context Protocol (MCP) server providing AI assistants with access to org-mode knowledge bases through search, content retrieval, and ID-based lookups
-
lutum-openai
OpenAI Responses API adapter for lutum
-
tower-llm
A Tower-based framework for building LLM & agent workflows in Rust
-
snapshell
snappy CLI that generates shell commands via OpenRouter LLMs
-
kimetsu-chat
Interactive REPL chat client + MCP sidecar + cross-harness bridge for kimetsu. Installs into Claude Code and Codex as a brain-managed memory layer.
-
tovli
— RuVector-based technical memory assistant
-
vamq
Consume audio chunk from Voice Activity Messaging via ZeroMQ
-
blazen-provider-atlascloud
Generated blazen-llm provider wrapper for AtlasCloud (atlascloud)
-
enzymeml-mcp
The EnzymeML MCP Server
-
atlas-knowledge-cli
Atlas CLI — compile, solve, decide, verify, and publish knowledge packages for AI agents
-
memora-mcp
MCP server for Memora: verifiable, citation-checked memory tools over stdio for Claude Desktop, Cursor, and other MCP clients
-
halter
Programmatic agent runtime, protocol, and tool harness
-
oramacore_lib
Internal structures and utilities used by OramaCore. See orama.com for more details.
-
pf_observability
PromptFleet observability facade (happy path)
-
babel
Provide Rust enums for Groq, SambaNova, Openrouter's llm model names
-
rubydex-mcp
MCP server exposing Rubydex semantic Ruby code intelligence tools
-
mollify-mcp
Model Context Protocol (stdio) server for Mollify — the one server every agent front-end calls
-
orcs-hook
Lifecycle hook definitions and configuration for ORCS
-
erinra
Memory MCP server for LLM coding assistants
-
bitrouter-substrate
Per-session ACP substrate: one session, one agent, turn queue + worktree
-
tt-auth
API key validation. Web auth (Auth.js) lives in the cloud repo, not here.
-
harnesssphere-ingest
HarnessSphere ingest plane: local OTLP receiver with host enrichment
-
cogito
Relentlessly generate AI slop using LLMs
-
ort_custom_op
writing custom operators for the onnxruntime in Rust
-
forge-agent
Forge — the model-agnostic AI coding agent. One harness, every provider.
-
clap-mcp
Enrich your CLI with MCP capabilities
-
agentix
Multi-provider LLM client for Rust — streaming, non-streaming, tool calls, MCP, DeepSeek, OpenAI, Anthropic, Gemini, Codex
-
assay-runner-core
Internal/experimental substrate for Assay measured-run workflows. Runner orchestration, archive assembly, and layer normalizers for the Assay-Runner candidate (Phase 2D Slice 2). No…
-
custom-utils
Provide customized tools based on features
-
codex-wrapper
A type-safe Codex CLI wrapper for Rust
-
rootcx-types
RootCX shared types
-
ferrum-kv
KV cache management with PagedAttention for Ferrum inference
-
brk_rmcp
Rust SDK for Model Context Protocol
-
llmtrace
Transparent proxy server for LLM API calls
-
ryu-dashboards
Home dashboards: a customizable, constantly-updating grid of widgets — an extracted Core capability crate (engine + SQLite store + widget-source resolution + refresh loop + device-image…
-
jutella-xmpp
XMPP – OpenAI API bridge
-
vibewindow-aisdk
An open-source Rust library for building AI-powered applications, inspired by the Vercel AI SDK. It provides a robust, type-safe, and easy-to-use interface for interacting with various Large Language Models (LLMs).
-
txtop
A fast terminal dashboard for LLM token usage, cost, and limits across providers
-
the-grid
An AI-powered agentic operating environment
-
solo-steward
Solo: consolidation pass (SWS dedup, REM integration, decay)
-
kedge-ledger
SQLite audit logging and deterministic replay of agent trajectories
-
aidapter
Unified AI provider adapter for Rust
-
garudust-cron
Cron-scheduled autonomous task runner for Garudust AI agents
-
cxpak
Spends CPU cycles so you don't spend tokens. The LLM gets a briefing packet instead of a flashlight in a dark room.
-
vibe-graph-git
Git fossilization layer for Vibe-Graph snapshots
-
trusty-gworkspace
Google Workspace MCP server for the Trusty suite
-
anyllm_proxy
HTTP proxy translating Anthropic Messages API to OpenAI Chat Completions
-
mcpr-tunnel
Tunnel client and self-hosted relay server for mcpr
-
llama-rs
A high-performance Rust implementation of llama.cpp - LLM inference engine with full GGUF support
-
openclaw-miniflux-mcp
MCP server for Miniflux RSS reader
-
tl-mcp
MCP (Model Context Protocol) integration for ThinkingLanguage
-
oxa-chatcompletions
OpenAI Chat Completions face of the oxa conversion library
-
qres_core
Deterministic no_std runtime for quantum-resilient swarm consensus with fixed-point arithmetic, Byzantine aggregation, and ZK proofs
-
yaak
Translate natural language to bash commands using an OpenAI-compatible LLM
-
kotonoha-core
Config, lesson loader, and CLI backend adapter for kotonoha
-
zerobox-windows-sandbox
Sandbox any command with file, network, and credential controls
-
mcd-mcp
Model Context Protocol server for Markdown CSV Document packages
-
rullama-mdap
Multi-Dimensional Adaptive Planning (MAKER voting framework) for the rullama agent framework — k-out-of-n voting, recursive decomposition, microagent dispatch, red-flag validation, scaling estimation
-
vein-recall
Shared memory system for AI agents and tools
-
robot_master_arena
Arena and tournament infrastructure for robot_master agents
-
diffsift-llm
LLM orchestration for DiffSift — Anthropic/OpenAI providers, prompt templates, pre-check filter
-
radkit
Rust AI Agent Development Kit
-
ecl-adapter-fs
Filesystem source adapter for ECL pipeline runner
-
guardrails-mcp-server
AI agent guardrails MCP server -- input validation, output filtering, policy enforcement, audit logging
-
axil-atlas-proto
Wire DTOs for the Atlas sync protocol — shared by the public axil-sync client and the (private) Atlas server so both agree on the protocol without code coupling
-
gs-hf
Fork of hf-hub. This crates aims ease the interaction with huggingface It aims to be compatible with the huggingface_hub python package, but only implements a smaller subset of functions.
-
mcplug
toolkit for discovering, calling, and composing MCP servers
-
mempalace-rs
High-performance, local AI memory with AAAK v3.2 protocol and temporal Knowledge Graph
-
rto-render
Renderers for Roteiro: docs site, Obsidian vault, and optional MCP server. Implementation detail of the roteiro CLI; no API stability guarantee.
-
ext4fs-cli
CLI and MCP server for ext4 forensic analysis
-
infra-compass-cli
Command line application to manage and query NREL's INFRA-COMPASS data
-
uselegion-automation
Self-hosted, multi-channel AI agent gateway and CLI
-
descry-tool-core
Core traits and types for descry-tool framework
-
zsfm-core
Shared traits, request/response JSON envelope, and config helpers used by every zsfm model crate
-
agentkit-capabilities
Capability abstractions shared by tools, MCP, and agentkit hosts
-
haste-fhir-generated-ops
Generated FHIR operations for Haste Health
-
tibba-llm
LLM client for tibba (OpenAI-compatible & Anthropic)
-
enya
Observability editor for humans, machines, and AI agents
-
triplets
A composable, deterministic text data pipeline for ML. Ingest, denoise, chunk, split, and sample multi-source corpora into reproducible training triplets.
-
sutra_afm_lib
Safe, high-performance Rust bindings for Apple's on-device foundation models (Apple Intelligence)
-
substrate-hub
Backend infrastructure for LLM orchestration - plexus, activations, storage
-
agent-context
Multi-backend agent context manager with three-zone memory model
-
solograph
Code intelligence: tree-sitter scanning, project maps, dependency parsing
-
ort-rs
Object Record Table - a CSV like structured data format with native object and array support
-
loomweave-cli
Loomweave: a code-archaeology tool that analyzes a codebase into a queryable entity/edge graph and serves it over MCP
-
policyai
mechanism for unstructured, composable policies that transform unstructured text into structured outputs
-
chat-rs
Build LLM clients with ease, attach them to your tools
-
algocline-engine
algocline Lua execution engine — VM, session, bridge
-
llmctx
LLM context builder
-
haskelujah-backend-wasm
WebAssembly backend for the Haskelujah Haskell compiler
-
papersmith
AI-Powered PDF Renamer in Rust
-
ggen-ai
Thin wrapper around genai for ggen - LLM integration with environment support
-
dllm
datalog LLM
-
optimus-cli
optimus — meta-skill for LLM/agent tooling (wiki, caveman, aura, graphify)
-
pakx-registry-client
pakx federated registry client — MCP Registry, Smithery, Glama, GitHub raw
-
rkg-core
Deterministic repository knowledge graphs and context infrastructure for AI-assisted software engineering
-
nannyd
Authorization and audit for AI agents that take real-world actions. CLI (nanny run) and Rust SDK (#[tool], #[rule], #[agent]).
-
gpp-timeline
gpp continuous timeline capture engine
-
toktier-cuda-driver
Minimal dynamically loaded CUDA Driver API host for TokTier's prebuilt kernels
-
llmctl
A CLI tool for testing and validating LLM (Large Language Model) services. Supports multiple providers including OpenAI, Gemini, Claude, Ollama, DeepSeek, and any OpenAI-compatible APIs.
-
arete-mcp
MCP (Model Context Protocol) server wrapping Arete streams for AI agent integration; runs as
a4 mcpinside the a4 CLI -
spreadsheet-read-mcp
An MCP server that lets LLM agents explore spreadsheet workbooks safely and deterministically
-
keelrun-cli
The
keelbinary: run | init | doctor | status | explain. The product’s face — every command has a byte-deterministic--jsontwin and stable exit codes (dx-spec §1–2, §5–6). -
chonkie
🦛 Chonkie, now in Rust 🦀: No-nonsense, ultra-fast, ultra-light chunking library
-
turbo-surf-page
turbo-surf tier 2: Page over the turbo-dom Rust parser + Navigator impl
-
savetracker
Watch game save directories and track changes in save files
-
shit
LLM-powered command correction
-
mcp-dashboard
A full-featured TUI dashboard for managing and monitoring MCP (Model Context Protocol) servers
-
dbmd-core
Reference library for db.md, the open standard for databases in plain files. Parsing, store walk, wiki-link graph, validation, query, and write-through indexes. Zero AI dependencies.
-
imbh-mcp
Model Context Protocol server for imbh: read-only telemetry tools over a transport-agnostic JSON-RPC dispatch, plus a stdio transport
-
llm-samplers
Token samplers for large language models
-
oxa-modelmap
The single, optional model-renaming injection point (spec/03)
-
nous-core
Core types, traits, and errors for the Nous metacognitive evaluation module
-
memory-crystal
Crystallized memory: lossy, reconstructive persistence for AI agents
-
rahti-mcp
Read-only Model Context Protocol server for Rahti projects
-
dioxionary
Remember all words in terminal!
-
qk-str
Common string utility functions as standalone CLI commands, for humans and AI agents
-
synaptic-store
Key-value storage for Synaptic agents: Store trait and InMemoryStore
-
codei
终端优先的 AI 编程 Agent，用自然语言在本地仓库中读代码、改代码、跑命令、调试问题。
-
myko-server
Myko server runtime — WebSocket, peer federation
-
claudectl-tui
Terminal UI, dashboard recording, and demo-mode fixtures for claudectl. Splits out of the binary so the TUI can evolve independently of the brain/coord/bus subsystems.
-
rustia-llm
Rustia-powered LLM tool-calling adapter for aisdk
-
foxtail
Status line for Claude Code users at Mozilla
-
codewhale-config
Config schema and precedence model for Codewhale
-
lmm-agent
Equation-based autonomous agent framework for the lmm ecosystem
-
ferrum-native-ops
Native operator ABI and artifact resolver for Ferrum
-
etalon-cli
Lightning-fast privacy compliance and tracking detection CLI
-
ecotokens
Token optimizer for AI coding assistants — hooks into Claude Code/Gemini CLI to compress shell outputs and track USD savings
-
deadband-core
Core orchestrator, policy engine, intervention, and replay for Deadband
-
arbor-graph-cli
Command-line interface for Arbor
-
api_xai
X.AI Grok API client for accessing large language models (LLMs)
-
gobby-core
Shared foundation primitives for Gobby CLI tools
-
aigateway
A unified AI Gateway for multiple LLM providers
-
shuttle-openai
Shuttle plugin for connecting to OpenAI
-
kalosm-language
A set of pretrained language models
-
teamctl
Declarative CLI for running persistent AI agent teams
-
clawft-tools
Tool implementations for clawft (file, shell, memory)
-
quiver-core
Domain types and traits for the quiver workspace
-
mecha10-ai-llm
Large Language Model integration for Mecha10 - Claude, GPT-4, Gemini, and local models
-
skrills-intelligence
Intelligent skill recommendations based on usage patterns and project context
-
kimun-notes
A terminal-based notes application
-
markitdown
designed to facilitate the conversion of various document formats into markdown text
-
crw-core
Core types, config, and error handling for the CRW web scraper
-
bamboo-memory
Memory storage and retrieval components for the Bamboo agent framework
-
aether-project
Project-local settings and agent discovery for the Aether AI agent framework
-
regenerator-core
Core library for Regenerator 2000: state, disassembler, analyzer, commands, parsers, and exporter for 6502/Commodore binaries
-
tokn-provider-zai
ZAI provider integration for tokn
-
eros-engine-llm
HTTP clients for OpenRouter (chat) and Voyage (embeddings), plus a TOML task→model config used by the eros-engine AI companion engine
-
agentop
A TUI process inspector for Claude Code and OpenAI Codex CLI — like top for AI coding agents
-
mi6
A top-like CLI for monitoring agentic coding sessions
-
openai-struct
利用openai的openapi生成的rust结构体
-
mink-core
Embeddable Rust runtime for the mink AI coding agent
-
spider_agent_types
Pure data types and constants for spider_agent automation. Zero heavy dependencies.
-
lash-restate
Restate durable execution adapter for Lash runtime effects
-
evolve-dashboard
Local-only web dashboard for Evolve
-
rullama-finetune
Cloud fine-tune APIs (OpenAI, Anthropic, Together, Fireworks, Anyscale, Bedrock, Vertex AI) for the rullama agent framework. Dataset pipelines live in
rullama-datasets; local LoRA/QLoRA/DoRA… -
sgr-agent-tools
14 reusable file-system tools for sgr-agent based AI agents
-
tokenmiser-config
Shared configuration types for TokenMiser
-
omp-llm-types
Canonical provider-independent LLM requests, events, media values, and capability traits
-
snippet-agent
Codebase parsing tool. It aggregates code (i.e. RAG) as an input to LLMs
-
claria-bedrock
Amazon Bedrock model calls for Claria: streaming Converse, token budgets, throttle retries, and forced-tool structured output
-
llm-kit-azure
Azure OpenAI provider for LLM Kit
-
files_to_llm
A CLI to parse your directory files contents to clipboard or an output file in order to simplify the work with LLMs
-
unified-entity-card
Unified Entity Card helpers for validating UEC JSON payloads
-
mcp-screenshot
MCP server for taking screenshots, listing windows, and monitors
-
alpine-ai
A lightweight, low-bulk interface for all major LLM providers
-
hirnd
hirn standalone daemon — gRPC, HTTP, and MCP server for cognitive memory
-
lumen-lsp
The AI-Native Programming Language
-
libra-leptos
Libra Design System — Leptos (Rust) financial components. 20 components ported from go-templ.
-
kotonoha-llm
Direct HTTP API backends for kotonoha (Gemini + OpenAI-compatible: OpenRouter / OpenAI / DeepSeek / custom base_url)
-
deepseeknova-agent
Agent implementations: main agent loop, coordinator, and plan-mode runners with tool use and streaming
-
mcp-supplier
Supplier MCP Server — a Supplier Relationship Management platform (suppliers, contacts, certifications & qualification, catalog & pricing, purchase orders, RFQ/sourcing, quality audits & SCARs…
-
ffs-budget
Token-budget aware filters, smart truncation, and footer-preserving trimming for ffs
-
ir-search
Local markdown semantic search with hybrid BM25+vector retrieval and LLM reranking
-
mdbook_rash
Binary to create doc from
rashcode -
vs-humanize
Human-like input synthesis for vibesurfer
-
agentool
Rust toolkit for AI agents: JSON Schema-defined tools for workspace files, search, web, Markdown, Git, memory, human-in-the-loop hooks, and todos
-
cleanr-agent
Natural-language agent and command palette for cleanr
-
shopify-mcp
MCP server for Shopify — products, orders, inventory, customers, webhooks
-
smith-mcp-config
config management for smith MCP server manager
-
ryu-monitors
Website monitors for Ryu (price / stock / keyword / content-diff / uptime): watch a URL on a schedule and alert on change, with notification fan-out (webhook / Telegram / Expo push / BYO SMTP email)…
-
baibot
A Matrix bot for using diffent capabilities (text-generation, text-to-speech, speech-to-text, image-generation, etc.) of AI / Large Language Models
-
bamboo-config
Configuration, settings, paths, encryption and keyword-masking for the Bamboo agent framework
-
cersei-provider
Provider trait and built-in LLM providers for the Cersei SDK
-
codewandler-flux-orchestrate
flux multi-agent orchestration: markdown agent roles, sub-agent spawner, plan-and-dispatch
-
llmptr-infra
Infrastructure layer for AI API protocol translation — model registry, thinking config, schema utilities
-
claus
An I/O less Anthropic API implementation
-
agentic-eval
Evaluate programs, CLI commands, programming languages, AI frameworks, and VM/sandbox systems for agentic AI use across four axes — token efficiency, determinism, reliability, and…
-
slippy-cli
AI Linter for Rust projects
-
quanttide-agent
量潮智能体工具箱 — 数据模型与 LLM 客户端
-
sai-cli
('sai') — Tell the shell what you want, not how to do it. Natural-language to safe shell command generator.
-
deectx
Local-first PII-masking reverse proxy for AI coding tools
-
shannon-nu-mcp
Modules to run a model context protocol (MCP) server that provides Nushell as a tool
-
microagents-cli
Batteries-included terminal-based agent built on top of the microagents framework
-
rivven-llm
LLM provider facade for Rivven — unified API for OpenAI, AWS Bedrock, and more
-
jikji-media-bridge
Optional media bridge API scaffolding for Jikji OCR and ASR integrations
-
agent-doc-element-queue
Queue element descriptor for agent-doc
-
anyllm_batch_engine
Batch orchestration engine with job queue, workers, and event-driven notifications
-
ferrus
AI agent orchestrator - drives a Supervisor-Executor workflow for autonomous software development
-
tinycortex
Rust core for the TinyCortex memory system
-
laminae-persona
Voice persona extraction and style enforcement for LLM output — learns how you write and keeps the AI on-voice
-
supercode-cli
Volter Harness — a lightweight, fully-customizable AI coding agent CLI in Rust. Any model via OpenRouter; natively continues Claude Code and Codex sessions.
-
memory-mcp
MCP server for semantic memory — pure-Rust embeddings, vector search, git-backed storage
-
search-mesh-core
Core scanning, tree-sitter AST probing, context squeezing, and patch primitives for Search-Mesh
-
kodegraf-mcp
MCP server exposing 17 code intelligence tools for Claude Code, Cursor, Windsurf, and Zed
-
socorro-cli
CLI tool for querying Mozilla's Socorro crash reporting system, optimized for LLM coding agents
-
voltiq-mcp
MCP server adapter (stdio + streamable HTTP) exposing voltiq as agent tools
-
harn-serve
Shared outbound workflow server core for Harn adapters
-
vibecode
Compile-time vibe coding
-
relayburn-sdk
Embedding API for relayburn — published to crates.io as the supported Rust surface
-
slickit
Semantic, LLM-Interpretable Component Kit — typed component system
-
sift-cli
Strip noise from error output. Powered by a local LLM.
-
microsandbox-server
microsandbox-serverimplements the sandbox server responsible for orchestrating sandboxes -
scouter-state
Holds global state for managing python and rust interop
-
think-and-ship
Unified MCP server for structured reasoning + execution tracking — one binary, four tool families
-
mii-memory
A local-first memory store for AI agents with CLI, MCP server, and web explorer interfaces
-
cargo-context-scrub
Standalone re-export of cargo-context's secret scrubber for non-Cargo codebases
-
opi-ai
Unified multi-provider LLM API with streaming support
-
protonmail-mcp-server
MCP server for IMAP email access via ProtonMail Bridge
-
crush-lint
AI-augmented linter for Crush — a thin hint layer over the real crush-frontend parser and semantic checks
-
oxen-cli
oxen CLI lets you add, commit, push, and diff terabytes of data with a git style interface. Version machine learning training data, model weights, game assets, and studio media. Diff CSV and Parquet files row by row…
-
aigw-openai
OpenAI provider for AI Gateway
-
dwctl
The Doubleword Control Layer - A self-hostable observability and analytics platform for LLM applications
-
mcp-protocol
Protocol definitions for the Model Context Protocol (MCP)
-
distributed_cli
distributedCLI for Distributed applications: contracts check/accept, scaffold projects, describe manifests, compile clients, and render schema artifacts. Also a library so other CLIs (e… -
lutum-claude
Anthropic Claude Messages API adapter for lutum
-
chat-completions
Generic OpenAI-compatible Chat Completions client for chat-rs. Bring your own server (Ollama, vLLM, llama.cpp, LiteLLM, etc.).
-
mplx
MPL CLI tools for registry management and validation
-
langchainrust
A LangChain-inspired framework for building LLM applications in Rust. Supports OpenAI, Agents, Tools, Memory, Chains, RAG, BM25, Hybrid Retrieval, LangGraph, HyDE, Reranking, MultiQuery, and native Function Calling.
-
gpt-image-2-skill
Agent-first GPT Image 2 CLI and installable skill runtime
-
turul-mcp-protocol-2025-06-18
Model Context Protocol (MCP) specification implementation - 2025-06-18 version
-
cerememory-adapter-openai
Cerememory LLM adapter for openai
-
lellm-provider
Provider adapters for LeLLM — OpenAI, Anthropic, and more
-
wgpu-llm-cli
Terminal-based chat interface for the wgpu LLM inference engine
-
lopi
High-performance Rust agent orchestrator for Claude Code
-
helmsman
Adaptive instruction server
-
nosy-cli
nosy: various contents summarization tool powered by artificial intelligence
-
terraphim_router
Unified routing engine for LLM and Agent providers
-
kq-config
Configuration parsing for kq (knowledge.toml): paths, providers, watcher and search settings
-
hanzo-code-exec
Python code execution for hanzo
-
egui-mcp-client
embed in egui apps for MCP integration
-
piko-permissions
Permission and access control for PikoClaw
-
aof-runtime
Agent execution runtime with task orchestration
-
car-inference
Local model inference for CAR — Candle backend with Qwen3 models
-
codex-extra-memory-mcp
MCP server binary for codex-extra-memory
-
aws-sdk-agentregistrycontrol
AWS SDK for Agent Registry Control
-
dynamo-config
Dynamo Inference Framework
-
memorywhale-cli
MemoryWhale CLI — local-first terminal memory. Record commands, sessions, and output into local SQLite; browse them in a built-in web dashboard. Nothing is uploaded.
-
tandem-orchestrator
Orchestrator models for Tandem missions and routines
-
ik-llama-cpp-2
ik_llama.cpp bindings for Rust (SOTA quants + MTP)
-
defect-llm
LLM provider implementations (Anthropic, Bedrock, OpenAI, DeepSeek) for the defect agent
-
rma-ai
AI-powered vulnerability detection for Qryon
-
tokenless
LLM token optimization CLI — schema/response compression, command rewriting, TOON encoding
-
face_id
Face detection (bounding boxes, eyes/nose/mouth markers), and facial recognition (embeddings)
-
semantex-core
Core library for semantex semantic code search (indexing, embeddings, search)
-
claude-rust
Interactive CLI for Claude AI
-
peon-core
The zero-trust engine driving the Peon workspace, providing dual-layer sandboxing and dynamic Casbin whitelisting
-
serdes-ai-agent
Agent implementation for serdes-ai
-
stynx-code-provider
Anthropic API provider with SSE streaming
-
scraps_libs
Scraps is a portable CLI knowledge hub for managing interconnected Markdown documentation with Wiki-link notation
-
ask_llm
request to whatever llm is the best these days, without hardcoding model/provider
-
deepstrike-core
Cross-language agent runtime kernel — pure computation, zero I/O
-
lago-api
HTTP REST API and SSE streaming for the Lago agent runtime
-
bouncy-mcp
MCP server exposing the bouncy headless browser as typed tools to LLM clients
-
thulp-mcp
MCP (Model Context Protocol) integration for thulp
-
awaken-tool-pattern
Glob and regex pattern matching for tool IDs in Awaken
-
blackwall-core
Core policy engine for the Blackwall Protocol — a deterministic execution firewall for AI agents
-
rwhisper
interface for Whisper transcription models in Rust
-
rolter-control
rolter control plane: management api and static ui host
-
forge-guardrails
Foundation types for an LLM-agent workflow framework
-
anyllm
Low-level, generic LLM provider abstraction for Rust
-
tersify
Universal LLM context compressor — pipe anything, get token-optimized output
-
shuru-sdk
Async Rust SDK for Shuru microVMs
-
aeonic-core
Core traits, types, and async runtime for Aeonic
-
forja
Lightweight AI Agent Engine Core Framework
-
rover-fetch
An MCP server for fetching and prepping web content for LLM agents
-
phenotype-mcp-asset
MCP server for asset build and discovery operations
-
exocortex-pack-mortgage-v1
The mortgage/lending ontology pack for Exocortex: 7 memory types, 5 entity types, 11 relationship kinds, and pack-local Crepe rules M1-M3. Composes beside dev-v1; authored against docs/ONTOLOGY_GUIDE.md.
-
llama-desktop
Desktop interface for Ollama
-
imece_core
IMECE v0.2.0 - Local-First Autonomous Agent Framework. Zero API dependency. Zero cloud lock-in. Full sovereignty.
-
memcontext
A reliable LLM context management engine
-
lutum-trace
Scoped in-memory tracing layer for span/event collection and testing
-
json_partial
Json fixing parser for imperfect json given by LLMs
-
sonr
High-performance semantic search tool for local codebases
-
agent-fn-registry
Registry of LLM agent tools: callable + schema + side-effects + defaults
-
gitrepomind
Predict and find related files in your Git history using logical coupling
-
arbor-watcher
File watching and incremental indexing for Arbor
-
cloakpipe-cli
The cloakpipe binary: self-hosted policy-enforcement gate for AI agent tool calls, with its evidence ledger and verifier
-
cpex
CPEX host facade: re-exports the praxis-policy engine with its bundled builtins
-
tact
Terminal interface for Nanocodex
-
motosan-workflow
DAG-based agent workflow runner with autonomous milestone orchestration
-
claude-code-history
Print the most recent Claude Code session for the current or selected workspace
-
rsigma-mcp
Model Context Protocol (MCP) server exposing the rsigma Sigma toolchain as tools and resources
-
rag-rat-oracle
Compiler-grade reference oracle for the rag-rat workspace: SCIP/LSP evidence ingestion, run manifests, and the verdict store
-
palnia-cli
CLI for Palnia productivity app
-
unigateway-protocol
Reusable protocol parsing and response rendering for UniGateway
-
meritocrab-llm
LLM evaluator trait and implementations for the Meritocrab reputation system
-
om-context
LLM context tool that scores project files by importance
-
precc-cli
CLI tool for PRECC: ingest sessions, manage skills, debug, report
-
pforge-runtime
Zero-boilerplate MCP server framework with EXTREME TDD methodology
-
kernex-skills
Skills.sh-compatible skill loader and trigger matcher for Kernex
-
cerememory-transport-grpc
gRPC transport binding for CMP
-
oven-sdk-azure
Azure OpenAI Chat Completions and Responses adapters for oven-sdk
-
mkcontext
that provides functionality for creating context
-
kimetsu-agent
Transport-neutral agent runtime for kimetsu: tools, prompts, verify-loop, providers (claude_code, anthropic, openai)
-
rtk-context-index
RTK — code graph indexing, symbols navigation, and blast radius impact analyzer
-
alex-store
SQLite trace store, pricing seed, and analytics for Alex
-
auxloclaw
Ultra-high-performance AI agent framework - beats Hermes and Nanobot
-
coreason-urn-authority
Epistemic Ledger & OCI Trust Anchor for CoReason URNs
-
plate-spinner
Dashboard for managing multiple concurrent Claude Code sessions
-
ecl-secrets
Pluggable secret resolution for ECL pipelines
-
ollana
Ollama over LAN - Auto-discover your Ollama server on your local network with hassle-free ease
-
chace
CHamal's AutoComplete Engine - An LLM based code completion engine
-
aicost
Compute LLM usage costs based on the genai::Usage type
-
aigw-gemini
Google Gemini provider for AI Gateway
-
execkit-mcp
MCP (stdio) server exposing execkit shell sessions to AI agents
-
jikji-core
Shared types, errors, path policy, and generated artifact constants for Jikji
-
stoa-capture
Stoa capture worker (transcript drain + PII redaction)
-
nexus-memory
Command-line interface for Nexus Memory System
-
pf_config
Layered configuration loader (JSON, env, dotenv, Cargo.toml section) with optional WASM-friendly path later
-
elkpi-oxa
Pure, in-process protocol-conversion library for OpenAI and Anthropic
-
trusty-embedderd
Unified ONNX embedding daemon with stdio + Unix-socket transports and BatchQueue (issue #164)
-
bamboo-skills
Agent skill management (definitions, store, selection, access control) for the Bamboo agent framework
-
littrs
A lightweight, embeddable Python sandbox for LLM tool execution
-
malware-modeler
Train logisitic regression models for benign vs. malicious files based on byte n-grams and published research, plus related tools.
-
agnosai
Provider-agnostic AI orchestration framework
-
mii-text
A small, unix-friendly CLI for talking to OpenAI-compatible LLM APIs
-
rinne-mcp
A thin, framework-free MCP (Model Context Protocol) client for Rinne: stdio + Streamable HTTP transports, initialize / tools/list / tools/call
-
agent-jj
Claude Code hooks for jj-colocated repos — git guard, workspace creation, and cleanup
-
fprovider
Provider library for the fiddlesticks agent harness framework
-
orca-core
Core types, configuration, Runtime trait, and error types for the orca orchestrator
-
kotonoha-tts
Kokoro TTS adapter for kotonoha — wraps
kokoro-en(pure-Rust, no espeak) and exposes WAV-bytes synthesis -
tiktoken-wasm
WASM bindings for the tiktoken BPE tokenizer
-
cortiq-core
CMF (Cortiq Model Format): a self-describing, memory-mappable binary container for quantized LLM weights, tokenizer, per-task masks and per-skill delta records
-
cageforge-command
Validated command, environment, and stdio requests for Rust sandboxes
-
forge-core-executors
Task executors for the Forge framework - handles task lifecycle and execution
-
resq-ai
Multi-provider LLM abstraction for ResQ developer tools
-
paip
like cat but through llm
-
tooltest
CLI conformance testing for MCP servers
-
cf-gears-mini-chat-sdk
SDK for mini-chat: policy plugin traits, models, and errors
-
pizen-lynx-parser
Internal source parser for Lynx: source -> SymbolIdentity + intra-file Relation
-
crux-model
Canonical model ID types and provider-specific parsers for crux
-
tiycore
Unified LLM API and stateful Agent runtime in Rust
-
secretenv-migrate
SecretEnv — alias-migration engine.
migrate_with_plan+build_migration_planlifted from secretenv-cli at v0.16 (B1 carry-forward from v0.15) so the MCP server can wrap the same library entry the CLI uses. -
yallm
Unified LLM API converter - proxy server
-
git-workflow-mcp
Git workflow MCP server for Olympian Agent pipeline
-
shore-config
Shared configuration types and TOML loader for the Silvershore chat daemon and its clients
-
poe2-agent
AI agent for Path of Exile 2 build analysis
-
codewhale-cli
Agentic terminal facade for open-source and open-weight coding models
-
rig-llama-cpp
Rig completion provider for local GGUF models via llama.cpp, with streaming, tool calling, reasoning, and multimodal (mtmd) support
-
nexus-protocol
Integration protocols for Nexus - REST, MCP, UMICP
-
mcp-janrain
An MCP server for Janrain (Akamai Identity Cloud)
-
agentic-evolve-mcp
MCP server for AgenticEvolve - pattern library for AI agents
-
rolter-auth
virtual keys, roles and access checks for rolter
-
mdx-rust-analysis
Rust source analysis, bundling, and safe editing for mdx-rust
-
llm-fallback-router
Multi-provider failover for LLM calls. Try Anthropic, fall back to OpenAI/Gemini/Bedrock on retryable errors. Per-attempt audit log. Zero runtime deps. BYO clients.
-
adk-rs
Rust port of the Google Agent Development Kit (ADK)
-
punch-arena
The Arena — HTTP API server for the Punch Agent Combat System
-
xet-core-structures
Core data structures including MerkleHash, metadata shards, and Xorb objects
-
agent-rules
Convention and validation for prescribed policy in AI agent instruction files
-
langfuse-openai
OpenAI integration for the Langfuse SDK
-
decy-llm
LLM context builder for Decy C-to-Rust transpiler
-
neleus-db
Content-addressed database in Rust for AI agent memory. Every write is a Merkle-DAG object with an offline proof, so you can show what an agent retrieved and when. Immutable history…
-
opendev-cli
Binary entry point for the OpenDev CLI
-
aha_openai_dive
fork openai_dive, ChatMessageContentPart add video. OpenAI Dive is an unofficial async Rust library that allows you to interact with the OpenAI API.
-
trusty-analyze
Sidecar code-analysis daemon for trusty-search: complexity, smells, quality, facts
-
mcp-scada
SCADA MCP Server — an industrial-control / SCADA platform for energy & utilities (sites & assets, tags/points telemetry, real-time readings + historian trends, alarm lifecycle, safety-interlocked…
-
githubclaw
Near-autonomous AI agents that manage open-source projects end-to-end using GitHub as the single source of truth
-
sema-notebook
Notebook interface for Sema — cell-based evaluation with HTTP API and browser UI
-
liter-llm-proxy
OpenAI-compatible LLM proxy server — model routing, virtual keys, rate limiting, cost tracking
-
zeph-tools
Tool executor trait with shell, web scrape, and composite executors for Zeph
-
objectiveai-sdk
ObjectiveAI SDK, definitions, and utilities
-
arct-ai
AI integration for Arc Academy Terminal - OpenAI, Anthropic, and local LLM support
-
ararajuba
A unified Rust SDK for working with large language models
-
memra-reference
Portable unfused ModelPlan reference executor for memra correctness gates
-
ospipe
RuVector-enhanced personal AI memory system integrating with Screenpipe
-
r8r
Agent-first Rust workflow automation engine
-
ainl-contracts
Shared policy contracts for repo intelligence, context freshness, and impact-first tooling (no OpenFang deps)
-
vs-protocol
Wire protocol for vibesurfer
-
veles-mcp
MCP (Model Context Protocol) server for Veles code search
-
converge-manifold-adapters
Generic storage, provider, and tool adapters for Converge contracts
-
rsai
Predictable development for unpredictable Gen-AI models. Let the compiler handle the chaos.
-
helm-sdk
Rust SDK for HELM — fail-closed tool calling for AI agents
-
pf_llm_client
Simplified, provider-agnostic LLM client built on protocol_transport_core
-
talos-provider
LLM client abstractions and provider-specific implementations
-
ai-sh
An AI command line assistant, which is context-aware and multi-turn capable
-
tiny-agent
一个小而完整的 Rust LLM Agent 运行时：可中断、可恢复、可观测、可插拔的 agent loop / A small but complete LLM agent runtime in Rust — an interruptible, resumable…
-
shuttle-mcp
Shuttle MCP server
-
golem-ai-llm-openrouter
working with OpenRouter APIs on Golem Cloud
-
plato-nervous
Room-specific model distillation for PLATO nervous system
-
quotachecker-tui
AI Agent Quota Monitoring TUI dashboard
-
lineprior-cli
CLI for building, evaluating, tuning, and querying explainable action-ranking prior books
-
tt-provider-anthropic
Anthropic provider adapter — the worked reference example. Auto-applies cache_control on system blocks >=1024 tokens.
-
mcp-ecsexpress-rust-aws
Hello World streaming http MCP
-
aonyx-agent
The agent with a real memory palace — Knowledge Graph + Hybrid Search + Time-machine. Agent loop + the
aonyxCLI. -
rskit-ai
Shared AI vocabulary for rskit AI/ML crates
-
meerkat-tools
Tool validation and dispatch for Meerkat
-
voltiq-analysis
Rules engine + hybrid AI analysis layer for voltiq
-
llm-tui-rs
Terminal UI for LLM chat with multi-provider support (Ollama, Claude, Bedrock)
-
evolve-llm
Minimal LLM client (Anthropic Haiku + Ollama) for occasional challenger generation
-
knot
Codebase Graph + Vector RAG Indexer for Java, TypeScript, JavaScript, Kotlin, Rust, Python, Groovy, C/C++, C#, Build Systems, and HTML/CSS codebases
-
slap-your-openclaw
Detect MacBook slaps and trigger OpenClaw roast events
-
llm_providers
A unified source of truth for LLM providers, models, pricing, and capabilities
-
openai-mcp-server
Generic OpenAI-compatible MCP server
-
ironclaw-agents
Agent patterns (ReAct, Router, Planner, Critic-Actor) for the IronClaw AI agent framework
-
fernwright-mcp
An open-source browser automation MCP server that controls your real browser tabs through a local extension
-
stkd-cli
CLI for Stkd - stacked diffs for Git
-
llama-cpp-sys-3
llama.cpp bindings
-
mcp-utils
Abstractions that extend rust-mcp-sdk for simplified MCP tool definition and server setup
-
lumen-vm
The AI-Native Programming Language
-
keel-enforce
Compile validation, error codes, and circuit breaker for keel
-
ruvector-solver-wasm
WASM bindings for RuVector sublinear-time solver
-
kwaai-inference
Inference engine for KwaaiNet - Candle-based ML runtime
-
devboy-jira
Jira provider for devboy-tools — IssueProvider/Provider implementation including project version CRUD over the Jira Cloud REST API
-
review_tools
Review tools for agentic-mcp: diff snapshots, lens-based review, pagination
-
typesec-cli
CLI for typesec — validate, check, generate, run
-
lc-providers
LLM provider integrations for langchainrust — OpenAI, Anthropic, Ollama, Gemini, etc
-
chio-cross-protocol
Shared cross-protocol bridge contracts and orchestrator runtime for Chio
-
synaptic-lancedb
LanceDB embedded vector store for Synaptic — serverless embedded vector database
-
agents-runtime
Async runtime orchestration for Rust deep agents
-
commitaura
Intelligent Git commit assistant with README integration
-
serbero
parties of a Mostro dispute resolve it themselves, and escalates to a human solver when needed
-
claude-rust-config
Configuration loading and settings management
-
ironflow-core
Rust workflow engine with Claude Code native agent support
-
quectoclaw
— Ultra-efficient AI assistant in Rust
-
unigateway-host
Reusable host contracts and compatibility surface for UniGateway embedders
-
ainl-context-compiler
LLM context-window assembly: multi-segment, role-aware, question-aware prompt orchestration for AINL hosts. Phase 6 of SELF_LEARNING_INTEGRATION_MAP. Distinct from
ainl-context-freshness… -
unrager-model
Shared data types for the unrager workspace (server, TUI, web/mobile clients)
-
drover
Prompt ollama like a unix tool- simple and scriptable
-
stt-cli
Speech to text Cli using Groq API and OpenAI API
-
clankers-ml
ML inference and model deployment for clankeRS
-
fabryk-mcp
Fabryk MCP — umbrella crate re-exporting all MCP components
-
mcp-cli-builder
Command-line interface generation for MCP servers built with mcp-utils
-
kcr_tempo_grafana_com
Kubernetes Custom Resource Bindings
-
avdoc
free cli for vibecoders
-
abu-rag
Retrieval-Augmented Generation system
-
helen-runtime
Helen language Rust reimplementation — helen-runtime (M5 LLM runtime)
-
rimz
The control room for your coding agents: a realtime tmux/Zellij dashboard for fleets of Claude Code, Codex, and Pi agents
-
silicon-waveform-client
Stateless Rust client for Silicon Waveform speech services, signed in with Silicon Accounts
-
allframe-mcp
MCP (Model Context Protocol) server for AllFrame - Expose APIs as LLM-callable tools
-
unistructgen
A powerful Rust code generator
-
toli
generating callable tools from functions to be used in Large Language Models (LLMs) or other systems, with JSON argument parsing and doc comment integration
-
typesec-agent
Agent executor with typestate and capability-based access control
-
pmcp-widget-utils
Shared widget utilities for PMCP SDK — bridge script injection
-
simple-agents-router
Routing strategies for SimpleAgents providers
-
clayers-spec
Spec-aware logic for clayers (validate, drift, coverage, connectivity)
-
bitrouter-sdk
HTTP client for the BitRouter service
-
latch-cache
Prompt cache metadata planning and injection helpers for Latch
-
agent-line
A batteries-included Rust library for building agent workflows
-
cera
Rust-native LLM inference engine
-
isimud-text-to-speech
AI-native macOS menu bar text-to-speech and MCP server for agents
-
modelstat
Privacy-first SDK for modelstat — wrap your backend LLM calls and ship redacted usage to a local daemon or the modelstat server, without touching live-request latency
-
mhome-agent-runtime
Canonical provider-neutral runtime and behavior for the MHome Agent
-
codineer-plugins
Plugin system and hooks for Codineer
-
loopctl
A trait-based framework for building agent loops with pluggable LLM clients, tools, and memory
-
rustdoc-mcp-server
MCP server for accessing Rust crate documentation
-
living-postmortem-mcp
MCP stdio server for incident postmortems stored as markdown in a vault
-
runtool
a.k.a. run: the bridge between human and AI tooling
-
rig-cerebras
Cerebras AI provider integration for the Rig AI framework
-
ras-llm-openai
Openai LLM adapter
-
ait
terminal based chat interface for interacting with large language models from various providers
-
orcs-types
Core types for ORCS CLI
-
lvz-protocol
Normalised contracts for the Lavoisier agent core: Provider, Event, Capabilities, Tool, Gateway, Tuner. Zero provider/gateway deps.
-
anthropic-proxy
High-performance Anthropic API proxy to OpenAI-compatible endpoints
-
silkai-server
HTTP and WebSocket daemon for SilkAI: OpenAI-style chat on top of the GPU scheduler
-
open-kioku-context
Context pack assembly for AI coding agents
-
harn-cli
CLI for the Harn programming language — run, test, REPL, format, and lint
-
pf-registry
ProcessFork registry adapters: file, Hugging Face Hub, S3-compatible, IPFS, local OCI
-
rho-providers
Model provider runtimes, credentials, and model catalog for Rho
-
dynamo-parsers
Reasoning and tool-calling parsers for OpenAI-compatible inference output
-
mcp-spec
Core types for Model Context Protocol
-
aether-telemetry
OpenTelemetry support for Aether GenAI calls
-
mcplease
mcp framework
-
scouter-drift
Drift logic for Scouter
-
fulfil-mcp
MCP server for Fulfil.io ERP — inventory, orders, shipments, customer management
-
tuitbot-mcp
MCP server for Tuitbot autonomous X growth assistant
-
revka
— memory-native AI agent runtime powered by Kumiho
-
otter-ai
Unified LLM API with provider collections, automatic auth resolution, token and cost tracking, and context persistence
-
wicked-estate-tree-sitter-rpg
Free-format RPG IV (ILE RPG) grammar for tree-sitter — wicked_estate symbols+calls subset
-
log_structured_store
A high-throughput append-oriented storage engine built in Rust using batched buffer-ring writes, direct I/O (O_DIRECT), and async io_uring dispatch
-
codeoid-protocol
Wire protocol types for the Codeoid daemon — serde-compatible Rust port of src/protocol/types.ts
-
mojentic
An LLM integration framework for Rust
-
graphify-cache
SHA256-based semantic caching for graphify
-
missiond-daemon
Singleton missiond daemon - task queue, PTY management, WebSocket API, and MCP IPC
-
vterm-rs
PTY orchestrator that lets AI agents drive real terminals
-
veles-core
Core library for Veles code search — indexing, chunking, BM25, dense search, and hybrid ranking
-
concrete-token-demo
Concrete Token JEPA demo using local Liquid AI models
-
codewandler-flux-core
flux core contracts: content blocks, messages, streaming chunks, events, errors (pure, no IO)
-
wwama
Rust wrapper crate for llama.cpp with native, wasm32, and wasm64 builds
-
hanzo-mcp-server
Hanzo MCP Server - Rust implementation with search, tools, and code analysis
-
aws-sdk-agentregistry
AWS SDK for Agent Registry
-
vex-hardware
Hardware-rooted identity for AI agents (TPM/CNG)
-
pf_mcp_protocol
Model Context Protocol (MCP) implementation for PromptFleet - JSON-RPC 2.0 compatible
-
rust-agent
Next Generation AI Agent Framework for Web3
-
nodetoken
KeyCompute node-token client for personal PC nodes
-
blazen-provider-perplexity
Generated blazen-llm provider wrapper for Perplexity (perplexity)
-
fleet-warden
Automated disk cleanup daemon for WSL development environments
-
org-cli
Command-line tool for searching, reading, and managing org-mode knowledge bases with fuzzy text search and structured content access
-
roder-tui
Agentic software development tools and SDKs for Roder
-
ccstat-mcp
MCP server for ccstat
-
reflect-mcp
MCP server for reflect
-
oxide-mcp-server
MCP (Model Context Protocol) server for Rust Oxide. Exposes oxide-gen generated CLIs and oxide-browser-sh / oxide-k bus actions as callable tools over stdio JSON-RPC.
-
claudix
Local semantic search plugin for Claude Code
-
rig-nvidia
NVIDIA NIM provider integration for the Rig AI framework
-
openai-openapi-http
Client for OpenAI API generated from OpenAPI spec
-
dociium
Multi-Language Documentation & Code MCP Server - Fast documentation access for Rust, Python, and Node.js
-
kvlar-proxy
MCP security proxy — intercepts, evaluates, and logs agent tool calls
-
archmap
Static architectural analysis of codebases
-
agntcy-shadi-agent-transport-slim
SLIM transport integration and stdio bridge support for SHADI
-
ollama-file-find-cli
A command-line interface for the Ollama file find library
-
cc-statusline
Lightweight statusline for Claude Code showing context usage and costs
-
toast-api
An unofficial CLI client and API server for Claude/Deepseek
-
shack-blackboard
MCP server providing a shared blackboard with publish/subscribe messaging for multi-agent coordination
-
cmdhub-skills
Plugin/skill system for CmdHub — extensible command discovery and execution strategies
-
deepseek-acp-adapter
Alpha ACP adapter server that exposes DeepSeek as an agent to ACP-capable editors (Zed, Neovim, etc.)
-
ai_gettext_translator
A command-line tool that uses OpenAI’s models to automatically translate
gettextmessages in source code and.pofiles. Thought for Phoenix projects. -
ras-llm
LLM port traits and chat message domain types
-
reposcry-graph
Core graph model for RepoScry, a local code review graph engine for AI agents
-
claude-rust-errors
Centralized error types for the claude-rust workspace
-
mercutio
IO-less MCP server library
-
nano-agent
Lightweight Rust framework for building AI agents
-
magic_hour
Magic Hour Rust SDK
-
typesec-odrl
ODRL policy engine for typesec — W3C digital rights language subset
-
misanthropy
An interface to the Anthropic API
-
selfware
Your personal AI workshop — software you own, software that lasts
-
rustchatui
A small Rust library and binary for serving a pre-built ChatGPT-like WebUI in any Rust project
-
forge-sandbox-worker
Isolated V8 sandbox worker process for Forge
-
incurs
A declarative CLI framework for Rust with typed commands, agent discovery, HTTP, and MCP
-
mcp-inventory
Inventory & Warehouse MCP server — stock management, BOM, locations, transfers, reorder alerts, lot tracking, reservations
-
baseplate
Shared support crate for the Barnett Studios agentic-harness toolkit — Promise-Theory verification value types, trace types, path resolution, a cxpak MCP client, YAML registry loading…
-
zeroclaw-memory
Memory backends, embeddings, consolidation, and retrieval for ZeroClaw
-
loomweave-llm
Loomweave LLM + embedding provider traits, concrete providers (OpenRouter / Codex CLI / Claude CLI), and the outbound HTTP/CLI transport for summaries and embeddings
-
assay-common
Common shared types and utilities for the Assay framework
-
bitrouter-attestation
Provider-agnostic confidential-inference verification (ConfidentialVerifier): client-side TEE attestation (L1) + exchange integrity (L1.5). First impl: NEAR AI Cloud.
-
awful_text_news
Summarize newspaper articles from sources that publish text-only or lite versions of their stories
-
poppins
Our mission is to be the default way developers create custom LLMs
-
stillo-mcp
MCP server for stillo terminal browser
-
ag-tui-text
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
runtara-ai
AI/LLM integration for runtara workflows — synchronous, ureq-based
-
toolhub-ingestion
Skill, plugin, MCP, GitHub, and session JSONL ingestion parsers for toolhub
-
agentic-core
Core AI orchestration library for Ruixen - the collaborative AI agent framework
-
ras-agent
Agent step loop, history, plan, rerun orchestration
-
shack-session-replay
MCP server that records and replays agent tool-call trajectories for deterministic regression testing
-
agent-proxy-rust-core
Core middleware trait, axum server engine, upstream forwarding, and auth for agent-proxy
-
katu-core
Core traits and types for the Katu AI Agent framework
-
ambi
A flexible, multi-backend, customizable AI agent framework, entirely based on Rust
-
agent-diva-providers
LLM provider integrations for agent-diva
-
llama-gguf
A high-performance Rust implementation of llama.cpp - LLM inference engine with full GGUF support
-
pai_mcp
MCP (Model Context Protocol) adapter for PAI-Kernel · exposes governance daemon API as MCP tools · reference integration
-
synaps-engine
Runtime engine — streaming, tools, MCP, skills, extensions, sidecar
-
cersei-mcp
MCP (Model Context Protocol) client for the Cersei SDK
-
rally-rs
Cross-agent messaging for CLI AI agents
-
object_detector
Object detection using ORT and the yoloe-26-seg model. This model can detect multiple objects per image, each having a tag, pixel-level mask, and a boundingbox. It's pretrained, it has a vocabulary of 4000+ objects.
-
fathomdb-embedder
FathomDB embedder runtime — built-in embedder implementations for the fathomdb-embedder-api trait
-
alex-wrap
Harness wrap layer for special capture (Amp and other non-base-URL clients)
-
harness-batch
Batch tool for AI agent harnesses — run shell commands across multiple directories with sequential/parallel modes, workspace fencing via realpath canonicalization, fail-fast, and per-target timeout
-
llama-crab
Safe, ergonomic and complete Rust bindings for llama.cpp
-
ckg-mcp
MCP server (JSON-RPC over stdio) exposing ckg queries
-
skl-cli
AI skills package manager — install, share and manage skills, contexts and agents across your team
-
astrid-llm
LLM provider abstraction with streaming support for Astrid
-
shuru-darwin
Virtualization.framework bindings for shuru
-
sparrow-core
Core event and runtime contracts for Sparrow
-
zerobox-utils-string
Sandbox any command with file, network, and credential controls
-
ryu-engines
Engine-agnostic inference-configuration primitive for Ryu: the per-request
SamplingConfig(temperature/top_p/top_k/min_p/penalties/mirostat/DRY/seed …) and per-launchLaunchConfig(context size… -
sim-lib-skill
Skill cards, policy records, browse metadata, and projections for SIM
-
devboy-clickup
ClickUp provider for devboy-tools — IssueProvider/Provider implementation against the ClickUp REST API
-
ggen-lsp-mcp
MCP server exposing ggen-lsp repair routes as a tool (leaf crate — avoids the ggen-core↔ggen-a2a-mcp cycle)
-
agentic-bin
Agentic unified CLI for configuration management
-
microagents-core
Core microagents framework library
-
google-ads-mcp
MCP server for Google Ads — campaign management, reporting, and optimization
-
openai-czh
OpenAI compatible API client library with streaming support
-
cchain
An AI-native modern cli automation tool built with Rust
-
opensession-parsers
Parsers for converting AI tool session data to HAIL format
-
yls-agi-rust-sdk
Rust SDK for the YLS AGI gateway with unified OpenAI, Claude, and Gemini clients
-
ingot-ir
Agent IR: the canonical, target-neutral representation of a compiled Ingot agent
-
kbolt
Local-first retrieval engine
-
jamjet-engram-server
Engram MCP server — memory layer for AI agents. MCP stdio + REST API.
-
agentic-forge-cli
CLI for AgenticForge blueprint engine
-
llm-orchestrator-secrets
Secret management for LLM Orchestrator with Vault, AWS Secrets Manager, and environment variable support
-
precc-core
Core library for PRECC: context resolution, skill matching, GDB, DB, mining, metrics
-
haskelujah-syntax
Green tree syntax and structural utilities for the Haskelujah compiler
-
zero-tui
Terminal UI widgets and app state for supervising ZERO
-
musu
OpenAI-compatible Rust client and Codex-backed adapter
-
gpt-image-2-core
Shared runtime, config, provider, and history core for GPT Image 2
-
tirea-extension-a2ui
Render interactive UI components to the frontend via AG-UI protocol for tirea
-
ai-agents-reasoning
Reasoning and reflection capabilities for AI Agents framework
-
nlproxy
High-performance LLM prompt compression, shielding, and firewall proxy core by IntelliDeep
-
spotlight-llm
A high-level Rust bridge for Apple's WWDC26 on-device LLMs and Core Spotlight RAG
-
blackwall-cli
CLI for the Blackwall Protocol — type
blackwalland you’re protected -
zeroclaw-gateway
HTTP/WebSocket gateway for ZeroClaw — REST API, web dashboard, webhooks
-
polyc-llm-openai
LlmProvider backend over the OpenAI-compatible /v1/chat/completions API (Ollama, LM Studio, vLLM, llama.cpp, OpenAI)
-
aa-proxy
Sidecar traffic interception proxy for Agent Assembly
-
ecl-sink-gcs
GCS sink stage for the ECL pipeline runner (error/rejected item output)
-
native-pyannote-rs
Speaker diarization using pyannote in Rust
-
lethe-core
Core retrieval library for lethe (Rust port). Hybrid BM25 + dense + cross-encoder rerank with clustered RIF.
-
plexus-substrate
Reference Plexus RPC server with conversation trees and LLM orchestration
-
yoclaw
Secure, single-binary AI agent orchestrator
-
synrepo
A context compiler for AI coding agents
-
tuff-core
Core lifecycle engine and adapter contract for Tuff
-
agent-doc-element-signals
Signals element descriptor for agent-doc
-
gemchat
Multi-agent Gemini chat CLI built with ratatui
-
gai-rs
Git commit message generator powered by Rust and AI of your choice
-
flowgentra-ai
building AI agents with graphs
-
ground-truth-cli-rust
A high-performance Rust reimplementation of the ground-truth-cli MCP server
-
kiki-sassy-desktop-announcer
Kiki — the sassy desktop announcer that refuses to be ignored
-
atomcode-tuix
Open-source terminal AI coding agent
-
dee-trends
Google Trends data as JSON from the CLI
-
ortools-sys
FFI bindings to Google OR-Tools C++ library
-
awsim-bedrock
AWSim Bedrock service emulator
-
locode-provider
Provider trait and API wires (Anthropic Messages, OpenAI Responses) for the locode coding agent
-
g3-planner
Fast-discovery planner for G3/GB AI coding agent
-
ccprof
Claude Code Profile Switcher - manage multiple user settings profiles
-
mi6-cli
CLI interface for mi6
-
codemov-cli
CLI for codemov — code intelligence for AI agents
-
siumai-provider-utils
Provider-facing utility seams for siumai
-
mnm-mcp
MCP JSON-RPC server over stdio and Streamable HTTP: thirteen tools plus the add_midnight_skills prompt
-
pensyve-mcp-tools
Shared MCP tool definitions for Pensyve servers
-
turul-mcp-protocol
Model Context Protocol (MCP) specification implementation - Current version alias
-
team-core
Shared library for teamctl: YAML schema, validation, and artifact rendering
-
ironclad-ai-agent
Secure-by-Design Autonomous AI Agent
-
svipall-solver
Captcha job store and HTTP API for svipall (in.php/res.php + createTask/getTaskResult)
-
chat-gpt-cli-rs
A terminal-first Rust chat client for OpenAI with streaming, modes, and conversation tooling
-
mecha-core
Provider-agnostic agent harness: loop, tools, MCP client, sessions
-
rsclaw
AI Agent Engine Compatible with OpenClaw
-
agent-doc-lease
TTL freshness primitives for agent-doc leases and sidecars
-
xet-data
Data processing pipeline for chunking, deduplication, and file reconstruction; used in the Hugging Face Xet client tools. Intended to be used through the API in the hf-xet package.
-
temporal-agent-rs
Durable AI agent execution on Temporal using AutoAgents traits
-
houston-memory
Character-bounded persistent memory for AI agents (Hermes model)
-
anno-lib
NER, coreference resolution, relation extraction, PII detection, and zero-shot entity types
-
llmy-client
All-in-one LLM utilities
-
zagens-runtime-adapters
Runtime tool/MCP/persist adapters for Zagens sidecar (D16 E1-a)
-
trans-epub
Translate EPUB with CLI
-
kumiho-construct
Construct — memory-native AI agent runtime powered by Kumiho
-
leakferret-cli
CLI for leakferret — context-aware secret detection with provider verification and ENV.fetch rewrites
-
rsrvr
Store all your LLM Interactions
-
orcs-lua
Lua scripting support for ORCS components
-
nexo-fork
Phase 80.19 — fork-with-cache-share subagent infrastructure (KAIROS port)
-
ragcli
CLI for local RAG
-
cerememory-store-emotional
Cerememory emotional store implementation
-
praxis-tools
Canonical tool implementations (filesystem, editing, shell, memory) for the Praxis tool engine
-
cognate-providers
OpenAI, Anthropic, and other LLM provider implementations for Cognate
-
silt
A transparent batching proxy for the OpenAI API that accumulates real-time requests and dispatches at intervals using the OpenAI Batch API to achieve ~50% cost savings
-
agentkit
Feature-gated umbrella crate for building agent applications with agentkit
-
tinkernel-client
Async Rust client for the Thinking Machines Lab Tinker API
-
symposium-ferris
Ferris MCP server - helpful tools for Rust development
-
alex-lar
Crash-recoverable, lazy indexed LLM archive for Alex
-
paddler
Open-source LLMOps platform for hosting and scaling AI in your own infrastructure
-
aeonic-providers
LLM provider adapters for Aeonic
-
forge-manifest
Hierarchical capability manifest for Code Mode tool discovery
-
mcpkit-rs
Rust SDK for Model Context Protocol
-
ras-telemetry
Anonymized telemetry events
-
ggml-sys-bleedingedge
Bleeding edge low-level bindings to GGML
-
synaptic-opensearch
OpenSearch vector store for Synaptic
-
llmman
Run any agent on any model, models stored as OCI images
-
await-openai
utils for openai
-
lutum-openrouter
OpenRouter adapter for lutum
-
baml-agent
Shared BAML agent runtime: config, registry, session persistence, loop detection
-
zag-serve
Network server for zag — remote access to AI agent orchestration
-
tokf-common
Shared types and utilities for the tokf CLI tool
-
homeboy-finding
Structured finding/diagnostic types for homeboy audit and review output
-
awaken-ext-observability
OpenTelemetry observability plugin with GenAI semantic conventions for Awaken
-
otelite-api
Lightweight web dashboard for visualizing OpenTelemetry logs, traces, and metrics
-
memex-contracts
Shared serde contracts for rust-memex transports and future frontends
-
llamactl
A command-line interface for managing and interacting with Ollama
-
punch-types
Core types for the Punch Agent Combat System
-
agent-framework
Microsoft Agent Framework: build AI agents and multi-agent workflows. Meta-crate re-exporting core and providers.
-
fastokens-b10
Fast Tokenizer
-
cf-gears-mini-chat
mini-chat gear: multi-tenant AI chat
-
routra-sdk
Routra Rust SDK - OpenAI-compatible client with intelligent GPU cloud routing
-
candor-daemon
Candor AI CLI and REST API daemon with voice, PDA, and 7-phase agent loop
-
cf-mini-chat
Mini-chat module: multi-tenant AI chat
-
cmdhub-mcp
MCP (Model Context Protocol) server for CmdHub — enables AI Agents to search and execute CLI commands
-
rv-tool
Non-invasive AI code review for any type of workflow
-
just-deepseek
DeepSeek API client and wire-level types
-
nousd
Nous daemon — metacognitive evaluation service
-
traiy_cli
serve AI suggestions according to user-provided guidelines and (optionally) context
-
nexo-assistant
Phase 80.15 — per-binding assistant-mode runtime
-
rullama-tool-runtime
execution runtime for the rullama agent framework — executor trait, registry, error taxonomy, sanitization, validation, transactions, smart router, plus optional orchestrator / OAuth…
-
vervet-technique
The Technique trait contract and self-registering inventory of emulation primitives
-
spadebox
Shovels for your AI agents
-
studio-worker
Pull-based image-generation worker for the minis.gg studio
-
tt-telemetry
OpenTelemetry setup + audit log primitives (hash-chain + Ed25519 signatures)
-
zed_llm_client
A client for interacting with the Zed LLM API
-
awful_aj
A CLI for interacting with OpenAI compatible APIs
-
httpcan
httpbin.org-compatible HTTP test server with OpenAI, Anthropic and OAuth 2.0 mocks
-
agent-diva-manager
Manager server for Agent Diva
-
kedge-policy
Lightweight user-space guardrails (blocked tools, PII redaction, budgets) from a TOML file
-
drift-mcp
Read-only stdio MCP server for Drift AI. Exposes blame / trace / log / rejected / show_event as tools.
-
nexo-tool-meta
Wire-shape types shared between the Nexo agent runtime and any third-party microapp that consumes its events
-
blazen-model-cache
Shared model download and cache layer for Blazen local-inference backends
-
caliban-tools-builtin
Built-in tools (Read/Write/Edit/Bash/Glob/Grep/WebFetch) for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
brainwires-telemetry
Unified telemetry for the Brainwires Agent Framework — analytics events, SQLite persistence, cost/usage queries, and a
BillingHooktrait consumed bybrainwires-billing -
rkg-cli
Deterministic repository knowledge graphs and context infrastructure for AI-assisted software engineering
-
tandem-observability
Tracing and observability utilities for Tandem
-
squick-dict
YAML pattern dictionary engine for Squick: framework and convention recognition without LLM calls
-
mcpi
A command-line tool REPL for inspecting MCP servers
-
rskit-agent
Agentic loop — Provider + Tools + Hooks in a turn-based execution engine
-
detect-coding-agent
Detect if your application is being invoked by an AI coding agent such as Claude Code, Copilot, Cursor, Codex, Gemini CLI, and many more
-
tabicl
— a tabular in-context learning foundation model (Rust port). Top-level prelude crate.
-
highflame-shield
Rust SDK for the highflame-shield guardrails service
-
arct-config
Configuration management for Arc Academy Terminal
-
sgr-agent
SGR LLM client + agent framework — structured output, function calling, agent loop, 3 agent variants
-
pensyve-mcp
Pensyve MCP server (stdio transport) — universal memory runtime for AI agents
-
stoa-recall
Stoa RecallBackend trait + reciprocal rank fusion
-
kodegen_candle_agent
KODEGEN.ᴀɪ: Memory-efficient, Blazing-Fast, MCP tools for code generation agents
-
solo-core
Solo: shared types and traits for the memory daemon
-
noether-engine
Noether composition engine: Lagrange graph AST, type checker, planner, executor, semantic index, LLM-backed composition agent
-
sysml-cli
CLI-native knowledge graph toolkit for SysML v2
-
microagents-events
Typed-, JSONRPC-compatible events definitions for the microagents framework
-
espionox
creating multi-agent workflows in Rust
-
sema-dap
Debug Adapter Protocol server for Sema
-
oxicode-vtui
Terminal UI framework for oxicode, based on vtcode-ui
-
sema-mcp
MCP (Model Context Protocol) server for Sema
-
ghidra-cli
Rust CLI to run Ghidra headless for reverse engineering with Claude Code and other agents
-
mcpr-widgets
CSP handling and HTML rewriting for MCP widget iframes
-
cc-agent-sdk
claude agent sdk
-
cageforge-path
Cross-platform native path validation for Rust sandbox policies
-
wave-dispatch
wave Dispatch — local-first AI router client. Route each request to the cheapest capable model (local-first; escalate to your frontier only when needed). BYO keys + infra.
-
blazen-local-llm
Shared option + request types for Blazen local LLM backends (candle, mistralrs, llama.cpp) and the controlplane local-model factory seam
-
mcp-execution-skill
SKILL.md generation for Claude Code integration with MCP tools
-
llm-optimizer-analyzer
Statistical analysis and anomaly detection
-
mcp-pos
Point of Sale MCP server — cart management, payments, receipts, shifts, loyalty, barcode lookup, returns
-
oneai-platform-desktop
OneAI desktop platform adapter — native UI approval gates for macOS, Windows, and Linux
-
hoard-rs
hoard - cli command organizer written in Rust
-
ras-tokens
Token cost service with LiteLLM pricing fetch
-
merlion-llm
LLM providers for Merlion Agent (OpenAI-compatible)
-
sim-codec-chat
Canonical chat transcript codec for SIM
-
graphify-serve
MCP server for graph queries
-
bamboo-server-tools
Framework-agnostic server-side tool implementations (memory, session inspector, skill runtime, compact, overlay) for the Bamboo agent framework
-
amore-mcp
Amore MCP server — Model Context Protocol stdio server exposing the Amore agent memory backbone (observe / recall / forget tools) to Claude Code, Claude Desktop, Cursor, Cline, Continue, and other MCP-aware IDEs
-
model-hf
Hugging Face model repository integration for Mesh LLM
-
crw-mcp-proto
MCP JSON-RPC 2.0 protocol types (shared by crw-server, crw-mcp, crw-browse)
-
rust-agent-client
LLM chat client (OpenAI / DeepSeek compatible) for rust-agent-framework
-
ecl
ECL workflow orchestration (part of Textrynum)
-
ferrum-types
Shared type definitions for the Ferrum LLM inference engine
-
ag-protocol
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
theway-transport
theway protocol layer — wire model + HTTP/WS/gRPC transports, independent of the server business logic, plus the shared client-contract modules (auth/config/feed/history/images/mentions/commands/triggers)
-
tirea-extension-observability
LLM inference and tool-call telemetry aligned with OpenTelemetry GenAI conventions
-
kalosm-language-model
A common interface for language models/transformers
-
oy-ai
AI provider abstraction layer for oy
-
shack-wasm-interpreter
MCP server that compiles and runs WebAssembly modules in an isolated, fuel-bounded sandbox
-
midstream
Real-time LLM streaming with inflight analysis
-
git-ca
git plugin that drafts commit messages using GitHub Copilot
-
ironclaw-providers
LLM provider implementations (Ollama, Anthropic, OpenAI, Groq) for IronClaw
-
gpt5_reasoner
GPT-5 prompt optimization and execution tool with MCP and CLI interfaces
-
etalon-core
Core vendor and context intelligence engine for ETALON
-
runifold-mcp
Capability-safe Model Context Protocol edge for Runifold
-
ras-types
Shared types and ID newtypes
-
ironclaw-cursor-brain
Use Cursor as an OpenAI-compatible LLM backend for Ironclaw
-
parari
A parallel AI agents runner
-
cargo-jarvis
Assistant conversationnel multi-IA avec interface HUD, bascule Gemini/Ollama et synthese vocale
-
axil-mcp
MCP (Model Context Protocol) server for Axil — JSON-RPC over stdio
-
g3-ensembles
Multi-agent ensemble orchestration for G3/GB AI coding agent
-
lmkit
Multi-provider AI API client (OpenAI, Anthropic, Google Gemini, Aliyun, Ollama, Zhipu; chat, embed incl. Gemini, rerank, image, audio stubs)
-
bamboo-sdk
Ergonomic top-level Agent SDK facade for the Bamboo agent framework
-
cortiq-cli
Command-line tools for the CMF model format: inspect, convert, run and serve .cmf models
-
gopher-mcp-rust
Rust SDK for Gopher Orch - AI Agent orchestration framework
-
nenjo-models
LLM provider trait and implementations — OpenAI, Anthropic, Gemini, OpenRouter, Ollama, and any OpenAI-compatible API
-
semver-analyzer-llm
LLM-based behavioral analysis for the semver-analyzer
-
thulp-registry
Tool registry and management for thulp
-
harn-vm
Async bytecode virtual machine for the Harn programming language
-
rustdoc-mcp
mcp server for rustdocs
-
iris-agent
A fast, token-efficient coding agent for the terminal
-
context-creator
High-performance CLI tool to convert codebases to Markdown for LLM context
-
rto-serve
Roteiro local model serving: an opt-in, loopback, OpenAI-compatible /v1 endpoint over installed models (ADR-0006). Implementation detail of the roteiro CLI; no API stability guarantee.
-
terraphim_spawner
Agent spawner for Terraphim with health checking and output capture
-
terse
Token compressor for LLM conversation history
-
ryu-rag
Retrieval-augmented-generation primitive for Ryu: the embedder (local hashing + remote OpenAI-compatible
/v1/embeddings), the reranker (local term-overlap + remote cross-encoder/rerank)… -
nuro
Unified public API for the Nuro AI agent SDK
-
cloudrun-test-rust
Basic Hello world Rust for tesing Cloud Run
-
ferriskey-cli
FerrisKey command-line client
-
tmux-claude-queue
A waiting-queue for multi-pane Claude Code + tmux workflows
-
repl-proto
JSON-RPC wire protocol types for the Symbi platform
-
agentic-mcp
Unified MCP server for all agentic-tools
-
tenere
TUI interface for LLMs written in Rust
-
swissknife-banking-sdk
Open Banking SDK for Rust - Plaid, TrueLayer, Teller, GoCardless, Yapily, MX
-
mem7-store
The core memory engine orchestrator for mem7
-
nexus-memory-mcp
MCP server implementation for Nexus Memory System
-
kreuzcrawl-cli
Command-line web crawler and scraper
-
edgecrab-acp
Agent Communication Protocol server for IDE integration
-
luckyshot
A CLI tool for one-shot code generations using RAG and file watching
-
punch-cli
CLI binary for Punch — The Agent Combat System
-
toktrie_tiktoken
tiktoken (OpenAI BPE) library support for toktrie and llguidance
-
nemo-relay-switchyard
First-party Switchyard Decision API routing plugin for NeMo Relay
-
cloakpipe-local
Local-only RAG mode for CloakPipe — future scope
-
nemo-flow
Core Rust SDK for NeMo Flow observability, scope management, and runtime instrumentation
-
onde-openai-harmony
OpenAI's response format for its open-weight model series gpt-oss
-
iwec
IWE MCP server: a Markdown knowledge base as structured memory for AI agents
-
haima-insurance
Agent insurance facilitation marketplace for the Agent OS
-
claude-rust-server
HTTP server with chat and health endpoints
-
magi-anthropic
Anthropic API client for Magi
-
gba-cli
CLI for GBA (Geektime Bootcamp Agent)
-
agentic-tools-registry
Unified tool registry aggregating all agentic-tools domain registries
-
zero-engine-client
Typed HTTP and WebSocket client for the ZERO paper engine
-
papers-core
CLI and MCP server for academic paper search, management, and local RAG
-
pay-respects-module-request-ai
AI request module for the pay-respects CLI tool
-
collet
Relentless agentic coding orchestrator with zero-drop agent loops
-
akar-ml
Local ML inference for Akar (ONNX embeddings)
-
ruchat
ollama/chroma command-line AI chat tool
-
codemode-rs
A secure V8 JavaScript sandbox with MCP (Model Context Protocol) tool integration
-
devboy-github
GitHub provider for devboy-tools — implements IssueProvider, MergeRequestProvider, and Provider against the GitHub REST API
-
puma
A lightweight, high-performance inference engine for local AI
-
aagt-providers
LLM provider implementations for the AAGT AI Agent framework
-
sparrow_agent
A full function Rust agent
-
codewhale-secrets
Secret storage backends for Codewhale, with OS keyring and file fallback, plus the shared pure secret-redaction and output-sanitization primitives
-
vterm-protocol
Wire protocol definitions for vterm-rs
-
pprog
An LLM pair programming server with web interface
-
ikigai-llm
Flexible LLM inference as ikigai ROC resources: urn:llm:ask (a facade grammar) dispatches to pluggable backends. Slice 0 ships an OpenAI-compatible REST backend covering Ollama / vLLM / llama…
-
assay-registry
Pack registry client for remote pack distribution (SPEC-Pack-Registry-v1)
-
omcp
functions
-
stakpak-api
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
nemo-relay-adaptive
Adaptive runtime primitives and Redis-backed learning components for NeMo Relay
-
golem-ai-llm-anthropic
working with Anthropic APIs on Golem Cloud
-
sim-lib-openai-server
OpenAI-shaped gateway routes and fixture media surfaces for SIM
-
shuru-proto
Shared wire protocol for shuru host/guest communication
-
actix_block_ai_crawling
Actix Middleware that blocks Generative AI crawlers
-
blazen-provider-deepseek
Generated blazen-llm provider wrapper for DeepSeek (deepseek)
-
paddle-ocr-rs
call Paddle OCR models via ONNX Runtime for image text recognition
-
lynx-protocol
Protocol definitions for Lynx
-
mcp-market-data
Market Data MCP Server — a market-data platform (instruments, real-time quotes and historical bars, analytics for returns/volatility/correlation/moving averages, yield & forward curves with interpolation…
-
codetether-agent
A2A-native AI coding agent for the CodeTether ecosystem
-
punch-memory
Persistence layer for Punch — SQLite-backed storage for fighters, bouts, memories, and knowledge
-
civx
client for the new CivitAI site API
-
zag-agent
Core library for zag — a unified interface for AI coding agents
-
deeplossless
Inference-aware execution runtime for long-running AI coding agents
-
pf-types
Lightweight naming and validation primitives for the PromptFleet agent ecosystem (AgentSlug, normalize_name)
-
sim-mcp-server
Final stdio and Streamable HTTP MCP server bootloader for SIM
-
aonyx-core
Aonyx Agent — shared core types, traits, errors
-
tuff-adapter-claude
Claude Code adapter for Tuff
-
agent-diva-neuron
Single-turn neuron node abstraction for agent-diva
-
roam-sdk
Roam Research SDK and terminal UI client
-
dexcost
Agent Unit Economics SDK for Rust — track end-to-end business-task costs for AI agents
-
serdes-ai-mcp
Model Context Protocol (MCP) implementation for serdes-ai
-
vkteams-bot
High-performance VK Teams Bot API toolkit with CLI and MCP server support
-
agdr-aki
Cryptographically-sealed AI decision records with provenance, audit, and compliance. Court-admissible inference logging for LLM accountability.
-
open-kioku-semantic-model
Shared semantic-model types and contracts for Open Kioku local code intelligence
-
vs-store
SQLite-backed durable state for vibesurfer
-
usls
integrated with ONNXRuntime, providing a collection of ML models
-
pomfret
Proxy Of Models For Routing, Evaluation & Telemetry — OpenAI-compatible API, backend routing, and web console
-
open-kioku-search-tantivy
Tantivy-backed full-text search for Open Kioku
-
forge-config
Configuration loading for Forgemax Code Mode MCP Gateway
-
arcan-commands
Slash command system for the Arcan agent runtime
-
openllm-rs
that enables seamless interaction with Llama.cpp backend services through an OpenAI-compatible REST APIe
-
icebreaker
A local AI chat app powered by 🦀Rust, 🧊iced, 🤗Hugging Face, and 🦙llama.cpp
-
merlion-agent
Merlion Agent CLI
-
ag-git
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
tally-ng
Git-backed findings tracker for AI coding agents
-
ai-req-proxy
ai-req-proxy是一个简单的 HTTP 请求转发代理服务器，专门用于调试和监控 AI API 请求
-
sim-lib-surface-card
Shared, plain-data surface-card spine
-
uira-orchestration
Agent definitions, SDK, tool registry, and hook implementations for Uira
-
cc-switch-tui
All-in-One Assistant for Claude Code, Codex, Gemini, OpenCode, OpenClaw & Hermes
-
stakpak-shell-tool-approvals
Shell command parsing and hierarchical scope resolution for tool approval
-
apiari-claude-sdk
Rust SDK for the Claude CLI — spawn, stream, and manage Claude agent sessions
-
fiat
A minimal Rust agent runtime for building custom AI agents
-
combs-formats
Combs Engine file-format adapters (ModelSource trait + safetensors)
-
vct-cli
Vibe Coding Tracker - AI coding assistant telemetry/usage CLI + TUI (the
vibe_coding_tracker/vctbinary) -
turul-mcp-json-rpc-server
Compatibility shim re-exporting turul-rpc — see turul-rpc on crates.io for current development
-
aof-mcp
Model Context Protocol (MCP) client implementation
-
codec-rs
Isomorphic tokenizer + detokenizer for the Codec binary transport protocol — for Rust. Decodes streaming token IDs from Codec-compliant servers (vLLM, SGLang) and encodes text into IDs for the bidirectional path.
-
vibe-analyzer
Agentic RAG engine for code and knowledge bases. AST extraction, LLM enrichment, OpenSearch indexing.
-
precc-shell
$SHELL wrapper that runs commands through the PRECC pipeline (for Aider and other tools without a hook surface)
-
ragit
git-like rag pipeline
-
terrain
A lightweight, configurable, local-first full-text search engine for Markdown knowledge bases with Japanese text support
-
airlock-rs
Local-first AI security gateway: PII redaction, token compression, and audit ledger
-
flash-rerank-cli
CLI for Flash-Rerank — compile, benchmark, serve, and download models
-
vw-shared
Shared utilities for VibeWindow crates
-
agntcy-shadi-a2a
A2A channel support for SHADI (SLIMRPC, gRPC, JSON-RPC, and HTTP+JSON)
-
apl-cpex
APL ↔ CPEX runtime bridge — per-hook PluginInvoker implementations
-
greentic-mcp-client
Client for remote MCP servers over Streamable HTTP (JSON-RPC 2.0)
-
stynx-code
— interactive AI coding assistant
-
candor-sentinel
Deterministic guardrails: 6 rules + 10 doctrine principles for agent safety
-
asw
一个通用的 Code Agent 工具配置切换器，支持将任意 OpenAI/Anthropic 协议模型接入到主流 Code Agent CLI 工具中
-
rig-helixdb
Rig vector store index integration for HelixDB
-
mcp-tokens
Analyze token usage of MCP servers
-
lucy-cli
A small local JSONL agent harness
-
scouter-evaluate
LLM Evaluation logic for Scouter
-
modelcli
Call any LLM from the command line via models.dev
-
ferrum-kernels
Unified compute kernels (CUDA/Metal/CPU) and model runner for Ferrum inference
-
cvkg
Cyber Viking Kvasir Graph (CVKG) - High-fidelity agentic UI framework
-
tt-shared
Shared types, errors, and Provider trait for TokenTrimmer
-
causality-mcp-server
MCP server binary for causality-mcp (Claude Desktop stdio JSON-RPC)
-
endpoints
A collection of data structures for the OpenAI-compatible endpoints
-
toolhub-core
Domain types and traits for the toolhub workspace
-
convergio-ceo
CEO routing engine — routes instructions to MCP tools via LLM
-
piko-config
Configuration management for PikoClaw
-
octoroute
OpenAI-compatible inference fabric for local pools and cloud providers
-
omni_search
A unified Rust SDK for multimodal embedding and similarity search
-
mnemefusion-llama-cpp-2
llama.cpp bindings for Rust (MnemeFusion fork)
-
cerememory-core
Core types, traits, and CMP protocol definitions for Cerememory
-
regenerator2000-tui
Terminal User Interface (TUI) for Regenerator 2000 using ratatui
-
sacp-proxy
Framework for building SACP proxy components
-
captcha-engine
ONNX-based captcha recognition engine
-
concerto-cli
Concerto CLI: a Rust inference multiplexer for self-hosted LLMs. Orchestrates vLLM, llama.cpp, and SGLang processes across 1-8 GPUs with demand-aware routing.
-
mcp-preview
MCP Apps Preview Server - Browser-based UI testing for MCP widgets
-
google-generative-ai-rs
An unofficial rust-based client library to interact with the Google Gemini generative AI API
-
rayclaw
Multi-channel agentic AI runtime with tool-calling, memory, and scheduling
-
zdev
Repository-backed coding guidance, task coordination, and independent review
-
argus-core
AI code review platform — your coding agent shouldn't grade its own homework
-
shore-llm
LLM provider abstraction and request orchestration for the Silvershore daemon
-
gitctx
MCP server for GitHub repository exploration
-
paca
Helpers for interacting with llama.cpp
-
fzp
Fuzzy Processor - parallel LLM inference pipe filter
-
swiftide-agents
Composable LLM agents, typed task graphs, and streaming RAG pipelines in Rust
-
codex-extra-memory
CLI for codex-extra-memory
-
dop-ensemble
Divergent–Originality–Precision Ensemble Framework for Generative Language Models
-
provnzero-proxy
A high-performance Zero Data Retention (ZDR) proxy for AI APIs
-
ingot-mcp
Model Context Protocol tool host for Ingot: serves an agent's declared tools from MCP servers over stdio
-
klaudes
CLI tool for managing multiple Anthropic API configurations
-
moe-llm-core
Part of the MoE-13 Ternary Intelligence Stack
-
shrimpl
programming language and LSP
-
vigy-mcp
MCP server for vigy — exposes runtime CRUD as MCP tools so Claude (and other MCP clients) can drive the registry
-
evolve-core
Core engine for Evolve: genome trait, operators, generation cycle
-
ctx-mcp
Local MCP server implementation for CTX
-
menta
Minimal Rust library for non-UI LLM and AI primitives
-
k8s-mcp
Kubernetes MCP Server - Feature complete as kubectl
-
mcp-procurement
Procurement MCP server — purchase orders, RFQs, supplier management, approvals, receiving, spend analytics, contract tracking
-
zerobox-network-proxy
Sandbox any command with file, network, and credential controls
-
gpt-image-2-web
Self-hosted Docker/Web server for GPT Image 2
-
golem-ai-embed-openai
working with OpenAI embeding APIs on Golem Cloud
-
liter-llm-ffi
Universal LLM API client with Rust-powered polyglot bindings
-
ryu-drafts
Drafts for Ryu: a durable outbox for unsent messages. Owns the whole spine out-of-process — the draft store, the trigger vocabulary, and the claim protocol that keeps two desktop…
-
fetchium-mcp
Fetchium MCP server — Model Context Protocol integration
-
zeph-llm
LLM provider abstraction with Ollama, Claude, OpenAI, and Candle backends
-
caliban-provider
Provider-neutral message IR and trait for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
redshank-core
Core domain, ports, application, and adapters for the Redshank investigation agent
-
skg-effects-core
Core effect execution traits and errors
-
evolve-cli
Evolve CLI — drop-in passive evolution for AI coding assistants
-
mermaid-fixer
Automatically detects and fixes syntax errors in Mermaid diagrams within Markdown files
-
zerobox-utils-rustls-provider
Sandbox any command with file, network, and credential controls
-
oneai-platform-ios
OneAI iOS platform adapter — UIAlertController approval gate via C callback bridge
-
node-token
KeyCompute node-token client for personal PC nodes
-
kind-openai
Highly opinionated OpenAI API wrapper crate. By Kindness Inc.
-
lc-mcp
MCP (Model Context Protocol) client for langchainrust
-
tt-plan-core
Plan replay engine — deterministic simulation with bootstrap confidence intervals
-
ffs-grep
File grepping logic for ffs
-
thulp-guidance
LLM guidance and orchestration for thulp
-
oxi-mnemopi
Local SQLite vector memory engine — Rust port of omp Mnemopi
-
tokn-provider-llama-cpp
llama.cpp OpenAI-compatible provider integration for tokn
-
nexo-webhook-receiver
Phase 80.12 — generic per-source webhook verification + event extraction primitives
-
wicked-estate-knowledge
Knowledge engine: typed relations, cite-able recall, cross-store code↔knowledge differentiator
-
heimdall-privacy-filter
OpenAI privacy-filter setup, ONNX Runtime inference, and text redaction for Heimdall
-
denyx-local-mcp
Denyx local-executor MCP server: routes a cloud orchestrator's step description to a local Ollama model that emits Starlark, then runs the program through denyx-mcp under policy
-
assay-metrics
Metrics library for Assay
-
gb-personas
✨💖 Theatrical personas for GB (G3-Glitter-Bomb), a fork of G3 dialectical autocoding 💖✨
-
nenjo-worker
Agent worker for the Nenjo platform
-
agent-error-recovery
Composable error recovery strategies for LLM agent tool calls
-
talos-tui
Terminal user interface for Talos agent
-
langgraph-providers
LLM providers integration (OpenAI, Anthropic, etc.) for LangGraph applications in Rust
-
starweaver-runtime
Agent-loop graph and runtime executor primitives for Starweaver
-
citadeldb-mcp
Model Context Protocol (MCP) server exposing Citadel encrypted memory as tools
-
agentic-planning-cli
CLI for agentic-planning
-
agentive
building agentic LLM applications with streaming, tool calling, and multi-turn conversation loops
-
ecl-pipeline-state
Pipeline execution state types and checkpoint logic for ECL pipeline runner
-
ai-agents-memory
Memory implementations for AI Agents framework
-
mcp-hris
HRIS MCP Server — employees, departments, time-off, payroll, org chart, and performance for AI agents
-
reviva
Local-first review terminal for deterministic, inspectable, constrained repository analysis
-
algocline-app
algocline application layer — execution orchestration, package management
-
cli_engineer
An autonomous CLI coding agent
-
journal-mcp-rmcp
rmcp (MCP) interface layer for journal-mcp: ServerHandler, tool_router, and stdio / streamable HTTP transport wiring over journal-mcp-core
-
defect-agent
Core agent runtime for defect: turn loop, context compaction, tools and session orchestration
-
contextd
A local-first, semantic context daemon for AI agents
-
ratatalk
A terminal chat client for Ollama, built with Rust and ratatui
-
devboy-core
Core traits, types, and error handling for devboy-tools — Provider, IssueProvider, MergeRequestProvider, configuration model
-
red-green-refactor
project to demonstrate the red-green-refactor cycle in TDD
-
cage-bro-runtime
Process runtime for cage-bro sandbox (PTY shell, Landlock filesystem isolation, rlimits)
-
cloakpipe-mcp
MCP tool-call gate for CloakPipe: contract checks, approvals, side-effect tracking and evidence on every tools/call
-
astrid-hooks
Hook system for Astrid secure agent runtime
-
synwire
Convenience re-exports and reference implementations for Synwire
-
gnaw-core
Core library for gnaw: code-ingestion primitives, wire DTOs, and the staged prompt-assembly pipeline
-
lc-prompts
Prompt templates for langchainrust
-
agentic-tools-core
Core traits and types for agentic-tools library family
-
zeph-orchestration
Task orchestration: DAG execution, failure propagation, and persistence for Zeph
-
kwaak
Run a team of autonomous agents on your code, right from your terminal
-
claude-rust-commands
Slash commands and file reference expansion
-
trace-weft-mcp
Model Context Protocol trace-context propagation for TraceWeft
-
symposium-rust-analyzer
Symposium rust-analyzer proxy - an MCP server connecting to rust-analyzer and a ACP Proxy managing it
-
sandhi-providers
Sandhi unified provider transport: adapter trait + Anthropic/OpenAI-compat/Gemini/Bedrock/Cohere/local adapters, with usage extracted at the source
-
golem-ai-stt-whisper
working with OpenAI Whisper transcription API on Golem Cloud
-
ffs-engine
Unified scanner, query dispatch, ranking and memory-budget enforcement for ffs
-
loong-contracts
Internal support crate for Loong: stable shared contracts
-
evolve-mutators
Mutation operators that vary AgentConfig to produce challengers
-
topodb-sgh
Structured Graph Harness — frozen-DAG agent execution over TopoDB
-
umi-memory
Memory library for AI agents with deterministic simulation testing
-
jutella
Chatbot API client library and CLI interface
-
memora-llm
Provider-agnostic LLM clients (Anthropic, OpenAI, Ollama) with a type-enforced redaction boundary, used by Memora
-
opencode-provider-manager
TUI/CLI binary crate for managing OpenCode provider configs
-
steer-workspace
Workspace trait and implementation for Steer
-
awaken-ext-skills
Skill package discovery and activation plugin for Awaken
-
byokey-store
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
nexo-llm
LLM provider clients (MiniMax, OpenAI-compat, Anthropic, Gemini) with rate limiter and tool registry
-
priest
Rust SDK for the priest protocol (spec v2.9.0)
-
tokn
Unified wrapper crate that re-exports the tokn Rust workspace crates
-
canon-core
Core data models and traits for Canon Protocol — verifiable, censorship-resistant knowledge sharing for AI
-
mcpr-session
MCP session management for mcpr
-
operonx
High-performance Rust execution backend for Operon workflows
-
xchecker-config
Configuration model, discovery, and validation for xchecker
-
mpl-proxy
MPL sidecar proxy for MCP/A2A traffic
-
greentic-desktop-runtime
Runtime host for Greentic desktop runner configuration, packages, MCP, and telemetry
-
oxllm
Minimalist adaptive routing LLM proxy in Rust
-
vct-tui
Terminal (ratatui / comfy-table) presentation layer for Vibe Coding Tracker
-
kalosm
interface for pretrained AI models
-
tandem-workflows
Workflow definitions and loading for Tandem
-
onetool
Sandboxed Lua REPL for LLM tool use
-
skillsrs-upstream
Upstream MCP server integration for skills.rs - Connect and aggregate external MCP servers
-
neomemx
A high-performance memory library for AI agents with semantic search
-
lellm-mcp
MCP (Model Context Protocol) Client / Server for LeLLM
-
hedos-gateway
The hedos gateway: a loopback server speaking the OpenAI and Ollama wire formats
-
agent-session-id
Generate and track session IDs for LLM agent conversations
-
genai-media
AI image and video generation CLI — fal.ai + OpenAI
-
nexus-memory-hooks
Agent hooks system for Nexus Memory System - automated memory extraction
-
brainwires-tool-runtime
execution runtime for the Brainwires Agent Framework — executor trait, registry, error taxonomy, sanitization, validation, transactions, smart router, plus optional orchestrator /…
-
cerememory-store-working
Cerememory working store implementation
-
llmy-clap
All-in-one LLM utilities
-
mem7-dedup
Memory deduplication logic for the mem7 memory engine
-
sapient-cli
Command-line tool and HTTP server for the SAPIENT inference engine
-
byokey-translate
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
stillo-llm
LLM bridge for stillo terminal browser
-
sysutils-stdiokey-rust
System Utilities with API Key MCP Stdio transport in Rust
-
arbor-server
WebSocket server implementing the Arbor Protocol
-
hallouminate
A markdown corpus indexer for LLMs to build and query their own per-repo wikis
-
optimus-aura
Cross-tool persistent memory + policy + verification (aura port)
-
ruvector-solver-node
Node.js NAPI bindings for RuVector sublinear-time solver
-
carryover
Zero-LLM-token context-handoff daemon — resume any AI session across Claude Code, Cursor, and Codex
-
claude-plugin-validate
CLI validator for Claude Code plugin manifests and plugin content schemas
-
openai_api_rust
A very simple Rust library for OpenAI API, free from complex async operations and redundant dependencies
-
mcp-lims
LIMS MCP Server — a Laboratory Information Management System (samples & chain of custody, tests/assays with specifications, results with out-of-specification detection and data-integrity-controlled amendments…
-
cerememory-decay
Memory decay engine - forgetting curves and noise accumulation
-
uira-tui
Ratatui-based terminal interface for Uira
-
ryu-teams
Agent teams for Ryu (a named, ordered collection of agents + a coordination strategy, addressed as one unit via @team): the SQLite store +
/api/teams/*HTTP surface, shipped as an… -
synaptic-e2b
E2B code execution sandbox for Synaptic
-
stynx-code-errors
Centralized error types for the stynx-code workspace
-
codewandler-flux-spec
flux tool/operation specs: JSON-schema I/O, effects, risk, idempotency, access, and pre-execution intent (pure)
-
typia-llm
Typia-powered LLM tool-calling adapter for aisdk
-
piko-tools
Built-in tools and utilities for PikoClaw agents
-
oxi-sdk
oxi AI agent SDK — build isolated, multi-agent AI systems
-
agentisc-relay-domain-policy
Control protocol FSM and message validation for the agentisc-relay
-
codewhale-agent
Model/provider registry and fallback strategy for Codewhale
-
optimus-llm
Provider-agnostic LLM cascade with retries + budget
-
smoleval
Minimal and lightweight evaluation framework for AI agents
-
sutegi-cli
The sutegi command-line tool: scaffold apps, models, and routes; introspect or REPL into a running app
-
clavem
Local proxy for the Anthropic Claude API that records per-model token usage
-
sapient-core
Core types, tensor, DType, buffer, and error handling for SAPIENT
-
simple-agent-type
Core types and traits for SimpleAgents
-
rag-rs
A Rust-native implementation of the RAG stack
-
toktrie_hf_downloader
HuggingFace Hub download library support for toktrie and llguidance
-
zust-llm
LLM request helpers for Zust runtime integrations
-
model-artifact
Model artifact selection, identity, and GGUF metadata helpers for Mesh LLM
-
b00t-c0re-a2a
A2A (Agent-to-Agent) Protocol Rust SDK core — AgentCard, Task lifecycle, Artifact/Message types, file-backed AgentStore, SkillRegistry. Foundation for agent interoperability per A2A v1.0 spec.
-
trace-weft-ingest
OTLP trace ingestion for TraceWeft that preserves original span IDs
-
cf-mini-chat-sdk
SDK for mini-chat: policy plugin traits, models, and errors
-
loong-kernel
Internal support crate for Loong: kernel primitives and governance core
-
varpulis-cluster
Distributed execution cluster for Varpulis streaming analytics
-
sabi-tui
Terminal AI agent with ReAct pattern - execute shell commands, run Python, analyze images
-
smos-domain
SMOS domain layer — entities, value objects, pure domain logic. NO IO dependencies.
-
intr-providers
Intentry model provider abstraction - Anthropic, OpenAI, Google, Ollama
-
mhome-agent-contract
Transport-neutral contracts and deployment ports for the MHome Agent
-
rkg-query
Deterministic repository knowledge graphs and context infrastructure for AI-assisted software engineering
-
ollama_for_pretty_graph
provide basic access to ollama api for systems based on pretty_graph
-
pleme-codesearch
Fast, local semantic code search powered by Rust — BM25, vector embeddings, tree-sitter AST
-
inferd-daemon
The inferd daemon: IPC over Unix socket / Windows named pipe (length-prefixed type-tagged frames for generation, NDJSON for embeddings), admission queue, single-instance lock, router, grammar support, activity log
-
kaiba-server
Kaiba API Server - HTTP API and infrastructure adapters
-
polaris_internal
Internal re-export layer for Polaris (use polaris-ai instead)
-
dynamo-backend-common
Shared runtime glue for Rust LLM backends
-
microagents-storage
Session storage layer for the microagents framework
-
liteforge
Rust SDK for LiteForge - LLM completions via OpenAI-compatible API
-
imaige-core
Async, HTTP-only Rust client for the Imaige AI generation API (image/video/text generation, async job model, SSE streaming)
-
tiy-core
Unified LLM API and stateful Agent runtime in Rust
-
oneai-skill
OneAI SKILL system with progressive disclosure and lightweight selector
-
nexil
Provider-agnostic LLM toolkit — streaming, tool calls, tape storage, OAuth
-
cerememory-store-episodic
Cerememory episodic store implementation
-
aonyx-api
Aonyx Agent — REST + WebSocket automation API over the agent core (Vague 4)
-
mnm-skills
Registry of embedded Agent Skill bundles plus harness detection and idempotent install logic shared by the CLI and MCP server
-
nexus-memory-llm
LLM provider abstraction for Nexus Memory System
-
mcp-execution-server
MCP server for progressive loading TypeScript code generation
-
paddler_download_manager
Open-source LLMOps platform for hosting and scaling AI in your own infrastructure
-
kapsl-core
Core model loading, configuration, and caching for the Kapsl inference engine
-
rig-surrealdb
SurrealDB vector store implementation for the rig framework
-
agent-doc-element-icebox
Icebox element descriptor for agent-doc
-
mini-agent
A minimal, extensible AI agent framework — async-first, multi-provider, tool-integrated
-
seher-claude-agent-sdk
Rust port of anthropics/claude-agent-sdk-python: drive the Claude Code CLI as an agent runtime
-
skillsrs-registry
Registry management for skills.rs - MCP server skill registration and discovery
-
ecl-adapter-slack
Slack stub source adapter for the ECL pipeline runner (validation)
-
tibet-store-mmu
MMU Illusion — userfaultfd proof-of-concept for transparent RAM virtualization
-
web-retrieval
Web fetch and web search MCP tools
-
todo-mcp
A Collaborative Todo List application and MCP server using CRDTs and Multicast
-
aether-core
Core library for AI-powered code injection framework
-
blazen-provider-cohere
Generated blazen-llm provider wrapper for Cohere (cohere)
-
unified-agent-api
Agent-agnostic facade and registry for wrapper backends
-
slm_llama
llama.cpp backend for slm_inference
-
devboy-gitlab
GitLab provider for devboy-tools — implements IssueProvider, MergeRequestProvider, and Provider against the GitLab REST API
-
sapient-ir
Intermediate representation, compute graph, op types, and optimisation passes for SAPIENT
-
firecrawl-sdk
Rust SDK for Firecrawl API
-
forja-memory
Memory and context storage provider for Forja framework
-
bonsai-pot
A from-scratch, dependency-light, dequant-free Bonsai (Qwen3-architecture) Q1_0 inference engine running on wgpu compute shaders
-
bamboo-mcp
Model Context Protocol (MCP) client — protocol, transports, executor, manager — for the Bamboo agent framework
-
mcp-sdk
Model Context Protocol (MCP)
-
synaptic-jina
Jina AI embeddings and reranker for Synaptic
-
mcplint-core
Core data models, adapters, and analysis framework for mcplint
-
llmg-gateway
HTTP gateway server for LLMG
-
yallm-ir
Intermediate Representation for yallm - unified LLM API converter
-
solo-api
Solo: MCP and HTTP transports
-
rig-memory
Conversations over time for the Rig agent framework: recent messages, compaction, and saving and resuming sessions
-
rllm
Unifies multiple LLM backends in Rust
-
querygraph
AI Navigator semantic layer for governed dataset discovery and reuse
-
daimon-provider-local
Locally-hosted model providers (Ollama, llama.cpp, llama-rs, OpenAI-compatible) for the Daimon AI agent framework
-
blazen
Rust-native AI workflow engine with event-driven orchestration
-
zeph-commands
Slash command registry, handler trait, and channel sink abstraction for Zeph
-
bamboo-compression
Compression utilities for Bamboo sessions and memory workflows
-
neuron
Composable async agentic AI framework for Rust
-
llm-agent
The agent library to build LLM applications that work with any LLM providers
-
slm_ikllama
ik_llama.cpp backend for slm_inference
-
model-ref
Model reference parsing and normalization helpers for Mesh LLM
-
ares-http
HTTP adapter plugin for ARES (Axum router, auth, overlay)
-
fierros-tools
Tool runtime and MCP integration primitives for Fierros
-
metalcraft-agent
Interactive AI coding agent with personas, skills, and tool approval
-
ras-judge
Judge eval for agent traces
-
ferrant
A lightweight, multi-provider AI agent framework in pure Rust (Agno-style)
-
llama-sampling
Sampling strategies for llama.rs — temperature, top-k/p, repetition penalty
-
agent-tool-allowlist
Allowlist/blocklist guard for LLM agent tool calls
-
papers-mcp
MCP server for academic paper search, management, and local RAG
-
mem7-vector
Vector index implementations for the mem7 memory engine
-
aegis-orchestrator
100monkeys.ai AEGIS orchestrator CLI and daemon
-
ferrum-quantization
Weight-format abstraction (Dense / GPTQ / AWQ / GGUF) for Ferrum models
-
grits-cli
Git-native, local-first issue tracker CLI and MCP server
-
synaptic
agent framework with LangChain-compatible architecture
-
zerobox-utils-absolute-path
Sandbox any command with file, network, and credential controls
-
zagens-cli
Zagens headless CLI + HTTP/SSE runtime sidecar (
zagens,zagens-runtimebinaries) -
synaptic-milvus
Milvus vector store for Synaptic
-
pearls
A lightweight CLI for managing long-running task graphs for coding agents
-
bitrouter-attestation-plugin
BitRouter attestation plugin: a RouteHook that tags (Record) or drops (Enforce) confidential routing targets by their TEE-attestation verdict
-
task-journal-core
Append-only event log + derived SQLite state for task-journal: schema, storage, pack assembler, classifier client
-
axiom_engine
Axiom-TTT Inference Engine – Rust implementation using candle
-
agentkit-adapter-completions
Generic OpenAI-compatible chat completions adapter for agentkit
-
pizen-lynx-storage
Internal storage layer for Lynx: SQLite graph/vector substrate plus a Tantivy lexical index
-
kvbm-common
Dynamo Inference Framework
-
loong-protocol
Internal support crate for Loong: protocol and transport types
-
stakpak-shared
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
navi-sdk
Local agentic engine and terminal-first coding agent
-
memra-sampling
Host-side sampler chain (temperature/top-k/top-p/penalties, llama.cpp CPU sampler semantics) for the memra inference engine
-
stateset-nsr
Neuro-Symbolic Reasoning (NSR) framework for hybrid AI systems combining neural networks with symbolic logic
-
fcommon
Common library for the fiddlesticks agent harness framework
-
kindling-client
Daemon-backed Rust SDK for kindling, a local-first memory engine for AI-assisted development
-
tldr-cli
CLI binary for TLDR code analysis tool
-
llm-kit-openai
OpenAI provider implementation for the LLM Kit
-
tencentcloud-hunyuan-sdk
TencentCloud Hunyuan SDK for Rust
-
latent-inspector
Fast CLI for inspecting and comparing learned representations across self-supervised vision models
-
open-kioku-parse
Source parsing and symbol extraction for Open Kioku
-
oxirs-chat
RAG chat API with LLM integration and natural language to SPARQL translation
-
punch-channels
Channel adapters for messaging platforms in the Punch Agent Combat System
-
forge-server
MCP server exposing search() and execute() Code Mode tools
-
agent-rate-limiter
Token bucket rate limiter for LLM agent API calls
-
pf_prometheus
Prometheus metrics integration for the PromptFleet observability stack
-
mlx-models
Model architectures and weight loading for mlx-server (LLaMA, Mistral, Qwen2/3)
-
kernex-adapter-core
Core Adapter trait and supporting types (AdapterId, Capability, Detection, AdapterRegistry) for Kernex agent adapters
-
rootcx-client
RootCX runtime client
-
llmy-types
All-in-one LLM utilities
-
mem7-graph
Graph store trait and providers for mem7
-
claude-list
CLI tool for viewing installed plugins, skills, agents, and MCP servers in Claude Code
-
ctxgraph-mcp
MCP server for ctxgraph — exposes graph memory to Claude and other LLMs
-
silkai-adapters
Engine adapters for SilkAI: llama.cpp in-process, llama-server/vLLM/Ollama over HTTP, and managed child processes
-
api_log_tracker
API log tracker with multi-provider LLM anomaly analysis, HTTP daemon, and TUI dashboard
-
tokenledger
Enterprise-grade token management and pricing governance system for AI coding agents
-
simple-agents-workflow
Workflow IR and validation for SimpleAgents
-
precc-hook
Claude Code PreToolUse:Bash hook binary for PRECC (< 5ms latency target)
-
llmsdk-provider
Provider trait abstractions for llmsdk (Rust port of @ai-sdk/provider v4)
-
siumai-provider-google-vertex
Anthropic Messages provider on Google Vertex AI for siumai
-
roder-ext-jsonl-thread-store
Agentic software development tools and SDKs for Roder
-
kvbm-config
KVBM Configuration Library for Tokio, Rayon, and Messenger runtimes
-
stateset-embedded-python
Python bindings for StateSet Embedded Commerce
-
shore-swp-server
Shore Wire Protocol (SWP) server implementation for the Silvershore daemon
-
car-mcp
MCP (Model Context Protocol) server library — transport-agnostic dispatch for exposing CAR capabilities. Used by car-mcp-server (stdio binary) and car-server (HTTP-streamable daemon endpoint).
-
openenv
Rust port of Hugging Face OpenEnv
-
deadband-adapter-langgraph
Deadband integration adapter for LangGraph
-
stowken-cli
Command-line interface for Stowken
-
gemini_bridge
Types and functions to interact with Gemini AI API
-
zeph-config
Pure-data configuration types for Zeph
-
ccs-proxy
Local logging reverse-proxy + dashboard for Claude Code / Codex traffic
-
zeroclaw-tools
Tool implementations for agent-callable capabilities
-
devboy-fireflies
Fireflies provider for devboy-tools — meeting transcripts and summaries over the Fireflies GraphQL API
-
lago-auth
JWT-based auth middleware for Lago — shared-secret validation, user context extraction, and session mapping
-
sparrow-memory
Persistent memory (SQLite facts, FTS5 search, knowledge graph, symbol index) and secret redaction for Sparrow
-
token-heat
LLM token usage monitor with heatmap visualization
-
ironclaw-config
TOML configuration and hot-reload for the IronClaw AI agent framework
-
lellm
Rust LLM orchestration framework — LangChain / LangGraph / AutoGen in Rust
-
hrdr-app
hrdr shared application core — UI-agnostic logic (slash-command registry, aliases, quit detection) shared by the TUI and headless runner frontends
-
siumai-provider-openai-compatible
OpenAI-like protocol standard for siumai (legacy crate name: *-compatible)
-
paw-rs
ProgramAsWeights Rust SDK — easy-to-use API for PAW inference
-
deepstrike-sdk
DeepStrike Rust SDK — agent framework built on deepstrike-core
-
sema-workflow
Sequential dynamic-workflow runtime + frozen JSONL run-directory journal for Sema
-
mcp-drone-imagery
Drone Imagery MCP Server — an aerial inspection platform (drones, flights/missions, geotagged imagery, inspection assets & spans with required clearance, detections with vegetation-clearance encroachment analysis…
-
agentsdk-openai
Autogenerated OpenAI types for agentsdk
-
axil-sync
Client seam for Atlas sync: selects the distillate from a local Axil and speaks the Atlas sync protocol. Fully standalone — Atlas is something you point it at, never a hard dependency.
-
pakx-core
pakx core — manifest, lockfile, resolver, installer logic
-
defect-cli
defect: a highly configurable, ACP-native, resource-frugal headless general-purpose agent CLI
-
cvkg-flow
Cyber Viking Kvasir Graph (CVKG) - High-fidelity agentic UI framework
-
oracle_omen_patches
Self-evolution patch system with gates for oracle.omen
-
rto-llama
Roteiro's llama.cpp inference core: the Engine trait + a llama.cpp-backed engine (generation, embeddings, vision), shared by serving and internal uses (ADR-0003/0006). Implementation…
-
agentic-server
Standalone axum server for agentic-api
-
lc-guardrails
Guardrails for langchainrust — input/output safety validation for agents
-
jamjet-models
JamJet model adapter layer — unified interface for LLM providers
-
raymon
Stateful MCP server and TUI for Ray-style logs
-
tokn-router-inspect
Loopback-only request and session history viewer for tokn
-
tldr-daemon
Background daemon for TLDR code analysis tool
-
swiftide-indexing
Composable LLM agents, typed task graphs, and streaming RAG pipelines in Rust
-
optimus-graphify
Folder to multi-modal knowledge graph (graphify port)
-
planecode
Ontology-based Rust coding agent. Indexes a workspace into a structural graph and drives an LLM through query/edit/verify loops.
-
dora-mcp-host
doragoal is to be a low latency, composable, and distributed data flow -
etalon-techscan
Blazing fast asynchronous technology fingerprinting engine
-
codecat
Merge a code repository into a single file (and split it back), respecting .gitignore. Ideal for LLM code analysis.
-
amsterdam
A gateway for observing, routing, and transforming prompts
-
openeruka
Self-hosted knowledge state memory server for AI agents — types, client, SQLite backend, REST API, MCP
-
sapient-generate
Text generation for SAPIENT — Pipeline::from_pretrained, KV cache, greedy/top-p/top-k sampling, streaming
-
codewandler-flux-agent
flux Agent pillar: agent definitions (AgentSpec, markdown Roles) assembled onto the flux-flow engine
-
cersei-hooks
Hook/middleware system for the Cersei SDK
-
navi-openai
Local agentic engine and terminal-first coding agent
-
lumen-runtime
The AI-Native Programming Language
-
convergio-kernel
Execution kernel — process lifecycle, sandbox isolation, resource limits
-
agentisc-relay-adapter-gemini
Gemini CLI adapter for the agentisc-relay using NDJSON over subprocess
-
edda-search-fts
Full-text search powered by Tantivy for Edda
-
awaken-ext-permission
Permission plugin with allow/deny/ask policies for Awaken tool execution
-
mcpr-events
Structured event system for MCP tool calls and sessions
-
oxios-kernel
Oxios kernel: supervisor, event bus, state store
-
topo-cli
CLI interface for Topo codebase indexer
-
nemo-fabric-core
Core config and runtime contract for NeMo Fabric
-
nvim-mcp-rust
Neovim AI assistant powered by Rust - Claude, OpenAI, Gemini, Ollama, LM Studio support
-
lethe-cli
Command-line interface for lethe; ships as the
lethebinary -
cloakpipe-proxy
OpenAI- and Anthropic-compatible LLM egress proxy for CloakPipe: PII pseudonymization in requests, rehydration in responses
-
agentbus-core
Core types for agentbus: envelopes and the spool store (registry, delivery, asks, event log)
-
project-map-cli-rust
High-performance architectural map generator for AI agents and developers
-
aeonic-gateway
HTTP gateway server for Aeonic — OpenAI-compatible API
-
linear-queries
Cynic queries and input types for Linear
-
reflect-eval
Evaluation runners and output parsers for reflect
-
reflect-core
Core types, traits, and pattern engine for reflect
-
aof-llm
Multi-provider LLM abstraction layer
-
egemma
Rust ONNX inference library for Google's EmbeddingGemma (text embeddings)
-
cerememory-association
Spreading activation engine for cross-modal memory association
-
llmtrace-security
Security analysis engines for prompt injection detection
-
brainwires-reasoning
Layer 3 reasoning — plan/output parsers + provider-agnostic local-inference scorers (router, complexity, relevance, retrieval, strategy, summariser, validator, entity enhancer) for the Brainwires Agent Framework
-
zeroclaw-channels
Channel implementations for messaging platform integrations
-
candor-core
Shared types: AgentState, IdealStateArtifact, AgentAction, CoreError
-
rmcp-agent
that extends langchain-rust with support for Model Context Protocol (MCP) tool integration and streaming tool execution capabilities
-
brainwires
Agent Framework — build any AI application in Rust
-
precc-cursor-hook
Cursor beforeShellExecution hook binary for PRECC
-
orcs-runtime
ORCS Runtime - Internal implementation (auth, channel, engine)
-
mcp-scheduling
Scheduling MCP server — appointments, shifts, resource booking, availability, recurring events, conflict detection, timezone-aware
-
navi-tui
Local agentic engine and terminal-first coding agent
-
beleth
Autonomous agent framework - The King commands legions
-
assay-runner-spike
Internal/experimental substrate for Assay measured-run workflows. Compatibility wrapper crate that re-exports the Assay-Runner candidate surface from assay-runner-schema (Slice 1) and assay-runner-core (Slice 2)…
-
zerch-core
Distance metrics for vector similarity search (cosine, euclidean)
-
aeonic-router
Intelligent routing engine for Aeonic
-
scouter-macro
Macro utilities for scouter
-
caliban-memory
File-backed memory tiers spliced into the caliban system prompt — internal crate for the caliban binary; no API stability, pin exact versions
-
aiproof-core
Core types (Document, Rule trait, Diagnostic, Severity, Span) for aiproof
-
rskit-tool
Tool definition, auto-wiring, registry and middleware for agentic systems
-
llm-daemon
LLM as a daemon
-
llama-mcp-server
Local LLM inference MCP server powered by llama.cpp
-
fastmemory
A Topology Ontological Clustering Engine for AI Agent Memory
-
cognis-trace
Pluggable observability for Cognis: bridges CallbackHandler events to Langfuse and (on the roadmap) OpenTelemetry
-
a2a-server-rust-azure
Minimal Rust A2A server agent
-
lanscope
Passive eBPF-powered IoT device fingerprinting & anomaly detection for your LAN
-
hallouminate-adapters
Adapters layer for hallouminate
-
crabot
A smart and powerful coding agent with native GUI
-
reflect-store
SQLite storage backend for reflect
-
flash-rerank-benchmarks
Benchmark suite for Flash-Rerank
-
agent-tool-timeout
Per-tool deadline enforcement for LLM agent tool calls
-
tasker-mcp
MCP server exposing Tasker capabilities to LLM agents and developer tools
-
llmsdk-xai
xAI provider for llmsdk (Rust port of @ai-sdk/xai)
-
mesh-llm-types
Shared protocol-facing data types for Mesh LLM crates
-
philharmonic-connector-impl-llm-openai-compat
OpenAI-compatible LLM connector implementation for Philharmonic, covering OpenAI, vLLM, and compatible gateways (llm_generate capability)
-
ag-orchestration
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
ogenius
Lightweight AI inference server using HuggingFace models directly - a simpler alternative to Ollama
-
clancy
A CLI wrapper that adds cross-session memory to Claude Code
-
oneai-core
OneAI core types, traits, and abstractions
-
ares-cordis
Cordis primitives for ARES: context, fiber, effect, and events. Leaf crate with no other ARES dependencies
-
tokn-config
Configuration loading and path resolution for tokn
-
pforge-config
Zero-boilerplate MCP server framework with EXTREME TDD methodology
-
piko-session
Session management for PikoClaw
-
sema-io
The process-wide tokio I/O pool behind sema-core's executor seam
-
greentic-pack
Greentic pack builder CLI
-
onnx-runtime-quantization
Shared audited quantization tables for ONNX Runtime execution providers
-
ai-agents
framework for building AI agents from YAML specifications with trait-based extensibility
-
agentic-aegis-core
Streaming validation, shadow execution, and security protection for AI-generated code
-
vibe-cluster
Multi-provider LLM cluster library for Rust
-
open_clip_inference
Run OpenCLIP compatible embedding models via ONNX Runtime
-
tandem-browser
Chromium browser sidecar and diagnostics for Tandem
-
agntcy-shadi-telemetry
OpenTelemetry initialization and tracing sinks for SHADI runtimes and tools
-
large
Rust LLM inference implementation
-
synaptic-slack
Slack message loader for Synaptic
-
graphify-security
URL, path, and label validation for graphify
-
assay-core
High-performance evaluation framework for LLM agents (Core)
-
skop
Skill manager for AI coding agents (Codex, Opencode, Antigravity)
-
simple_llama_rs
Ollama API
-
roder-ext-cursor
Agentic software development tools and SDKs for Roder
-
agent-framework-declarative
Declarative agents and workflows loaded from YAML/JSON specs
-
gpt-feeder
A command-line application that scans the entire codebase, and produces one string consisting of all filenames and file contents that you want included
-
rskit-llm
LLM provider abstractions for OpenAI and Anthropic
-
sema-docs
Canonical structured documentation for Sema builtins/special forms; powers LSP hover/completion and REPL apropos
-
tnk
Per-project sandbox VMs for AI agent runtimes
-
varpulis-mcp
MCP server exposing Varpulis CEP engine to AI agents
-
bamboo-storage
Session storage backends for the Bamboo agent framework
-
needle-c
C ABI shared library for Needle (FFI for C, Go, Swift, Python ctypes)
-
shore-swp-client
Async client connection library for the Silvershore daemon's Shore Wire Protocol (SWP)
-
agent-doc-element-status
Status element descriptor for agent-doc
-
atomcode
Open-source terminal AI coding agent
-
halter-providers
Providers crate for halter
-
ryu-vad
Voice activity detection (VAD) modality primitive for Ryu:
detect(frame) -> speech_probfeeding an endpointing/barge-in state machine. An extracted Core capability crate — in-process FOREVER (per-frame hot path… -
leone-gguf
Checked GGUF parsing and quantized block decoding for Leone
-
lancor
End-to-end llama.cpp toolkit: API client, HuggingFace Hub, server orchestration, benchmarks
-
ffs-c
Raw C API for ffs file finder
-
cerememory-adapter-gemini
Cerememory LLM adapter for gemini
-
tandem-core
Core types and helpers for the Tandem engine
-
assay-mcp-server
MCP server integration for Assay
-
forge-sandbox
V8 sandbox for executing LLM-generated JavaScript via deno_core
-
tt-provider-openai
OpenAI provider adapter — chat completions, embeddings, streaming
-
caliban-provider-anthropic
Anthropic Claude schema family for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
agent-event-log
Structured lifecycle event log for AI agent runs
-
funera
Flexible and security-oriented LLM agent framework for Rust — ReAct loop, tools, skills, middleware, security
-
lunar-chess
A screen overlay that detects chessboards, recognizes pieces using ONNX, and displays best moves from a chess engine
-
terraphim_multi_agent
Multi-agent system for Terraphim built on roles with rust-genai integration
-
gitgraph-cli
GitGraph terminal app (TUI + CLI) with commit description generation
-
stockpot-core
Core library for Stockpot AI coding assistant
-
lilyco-cli
Lilyco CLI renderer — generates clap::Command from CommandSchema
-
tokc
Count tokens from stdin or files, with colorized tokenization display
-
build-graph
Build a Rust code knowledge graph from build artifacts; includes a build-script helper and the
cargo build-graphCLI -
zqa-rag
A retrieval-augmented generation (RAG) crate, used by
zqa -
pizen-lynx-protocol
Internal frozen domain contracts shared across the Lynx crates
-
tt-mcp
Model Context Protocol server exposing TokenTrimmer intelligence to MCP-compatible clients
-
ncp-mcp-server
MCP adapter for NCP — exposes auditable WASM Brick graphs as MCP tools
-
opentslm
using Burn, WGPU, and llama.cpp
-
entelix
— web-service-native Rust agentic AI SDK (facade re-export crate)
-
ryu-model-catalog
Hugging Face model catalog + device-fit verdict primitive for Ryu: HF search/detail, GGUF tree inspection, per-node device detection and fit estimation, installed-model tracking, capability overrides…
-
lc-evaluation
Evaluation framework for langchainrust — Evaluator trait, built-in evaluators, dataset runner
-
deadband-proxy
Deadband Proxy — HTTP proxy that protects AI agents from loops
-
hanzo-agents
Specialized AI agents for development workflows - architect, CTO, code reviewer
-
aeonic-policy
Policy engine for Aeonic — routing rules, cost ceilings, rate limits
-
openai-api-types
Rust type definitions for the OpenAI API (chat completions, models, responses endpoints)
-
uira-providers
Multi-provider model clients (Anthropic, OpenAI, Gemini, Ollama) with OAuth authentication
-
runifold-providers
Feature-gated model provider adapters for Runifold
-
cortiq-net
Network pipeline-split for CMF: the worker wire protocol and the coordinator that drives layer spans across nodes
-
agent-sdk-provider
Optional provider adapter helpers layered over agent-sdk-core ProviderAdapter contracts
-
potato-head
Potato brands
-
traitclaw
AI Agent Framework — Simple by default, powerful when needed
-
pagedelta
Deltas: LLAMA's primary units of data
-
oomllama
Sovereign AI Inference Engine — .oom quantization, Trust Kernel encryption, JIS identity routing, TIBET provenance. The auto where trust-kernel is the motor.
-
agileplus-mcp-intent
Typed intent graphs for engineering work — convert prompts into validated DAGs of intents, features, tasks, tests, and edges, then store, query, and serve them from Rust, HTTP, or MCP
-
rolter-store
repository traits and storage backends for rolter
-
cmdhub-cli
cmdh — the CmdHub CLI client for offline command search and execution
-
astrid-cli
Command-line interface for Astrid secure agent runtime
-
cognate-rag
Retrieval-Augmented Generation pipeline and vector store integrations for Cognate
-
terraphim_task_decomposition
Knowledge graph-based task decomposition system for intelligent task analysis and execution planning
-
llama_cpp
High-level bindings to llama.cpp with a focus on just being really, really easy to use
-
kmp-application
Application services behind KMP's ten memory moves and evidence-backed recall
-
world-models-gpc-cli
CLI interface for GPC operations
-
uira-mcp-server
MCP server with LSP and AST-grep tools for Uira
-
devsper-providers
Distributed AI Swarm Runtime
-
hallouminate-daemon
Daemon layer for hallouminate
-
tirea-extension-interaction
Interaction extension for tirea
-
pawan-aegis
integration — generate pawan.toml from aegis manifests
-
devboy-slack
Slack provider for devboy-tools — MessengerProvider/Provider implementation against the Slack Web API
-
simple-agents-core
Core SimpleAgents client API integrating providers and healing
-
codewhale-command-contract
Prototype command capability and dispatch shapes for the staged TUI command extraction
-
shuru-vm
High-level VM management for shuru
-
mem7-llm
LLM client abstraction for the mem7 memory engine
-
vs-daemon
Daemon that owns the browser engine for vibesurfer
-
icvs
InstructCanvas - DAG-based instruction format for agentic AI tools
-
tftio-silent-critic
Silent Critic: a supervision framework for agentic software development
-
ecl-adapter-sftp
SFTP source adapter for the ECL pipeline runner
-
kcode-kennedy-session-media-contracts
Effect-free Kennedy session media validation
-
llama-kv
KV cache management for llama.rs — prefill, decode, paging, eviction
-
open-kioku-languages
Language detection and metadata for Open Kioku
-
aleph-alpha-client
Interact with large language models provided by the Aleph Alpha API in Rust code
-
punch-skills
Skill/move system for the Punch Agent Combat System
-
cortex-mem-rig
Rig framework integration for Cortex Memory V2
-
aeonic-state
Distributed state, memory and vector store for Aeonic agents
-
hotl-provider-anthropic
Internal component of hotl - no semver promise; pin exact or don't depend. Anthropic Messages API provider: SSE streaming, verbatim block assembly, static cache placement.
-
agentic-tools-mcp
MCP server integration for agentic-tools library family
-
cera-ffi
UniFFI bindings for cera — exposes the core inference engine to Kotlin, Swift, Python, etc
-
rust-code
AI-powered TUI coding agent written in Rust
-
cpex-plugin-audit-logger
CPEX CMF plugin — structured per-request audit logging
-
sema-llm
LLM provider integrations (Anthropic, OpenAI) for the Sema programming language
-
opsis-core
Core types and traits for the Opsis world state engine
-
agentic_logging
Centralized JSONL logging infrastructure for agentic tools
-
synaptic-nomic
Nomic AI embeddings for Synaptic
-
rewind-cn
Autonomous coding agent orchestrator built on CQRS + Event Sourcing
-
llmy-agent-tools
All-in-one LLM utilities
-
zeroclaw-runtime
Miscellaneous ZeroClaw subsystems: security, observability, gateway, cron, SOP, skills, hardware, TUI, and more
-
chatkit
Thin async Rust wrapper around any OpenAI-compatible chat API with streaming, tool calling, and structured output support
-
jammi-ai
Embeddable AI engine for inference, embeddings, vector search, and fine-tuning
-
tandem-ai
Tandem engine service and CLI binary
-
rkg-lang-rust
Deterministic repository knowledge graphs and context infrastructure for AI-assisted software engineering
-
infer-cli
CLI for the infer AI agent platform
-
cersei-lsp
LSP client for the Cersei coding agent SDK
-
ras-llm-openrouter
Openrouter LLM adapter
-
ras-skills
Browser-use Skills service client
-
zsfm-chronos
Chronos-2 (Amazon) zero-shot time-series forecaster: GGUF conversion + Burn inference
-
praxis-skills
SKILL.md discovery and activation for the Praxis tool engine
-
ag-harness
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
mhome-agent-protocol
Typed cloud wire protocol for the MHome Agent
-
tauri-plugin-device-ai-apis
A Tauri plugin for cross-platform device AI capabilities including speech, vision, and text processing
-
codescout
High-performance coding agent toolkit MCP server
-
claude-insight
Local observability and audit tooling for Claude Code sessions
-
llm-optimizer-actuator
Action execution and system orchestration
-
agentflare-gateway-registry
Gateway registry — intent-first discovery and execution across MCP servers
-
codescout-embed
Shared embedding primitives for codescout + librarian-mcp
-
qex-mcp
Lightweight MCP server for semantic code search — BM25 + optional dense vectors + tree-sitter chunking
-
forgekit-agent
ForgeKit agent layer - Workflow orchestration and AI loop
-
mc-tools
Tool system for magic-code AI coding agent (30 built-in tools)
-
cctakt
TUI orchestrator for multiple Claude Code agents using Git Worktree
-
fpo-rust
Fast license plate OCR inference in pure Rust - a port of fast-plate-ocr with ONNX model support
-
rig-fastembed
Rig vector store index integration for Fastembed. https://github.com/Anush008/fastembed-rs
-
trace-weft-otel
OpenTelemetry export bridge for TraceWeft traces
-
agnt
A dense, sync-first Rust agent engine — multi-backend inference, parallel tool dispatch, SQLite persistence, streaming. Flagship meta-crate.
-
agent-team-mail-daemon
Plugin daemon for agent-team-mail: GitHub issues, CI monitoring, cross-computer bridge
-
modelexpress-common
Shared utilities for Model Express client and server
-
whipplescript-kernel
Runtime kernel and provider integrations for WhippleScript workflows
-
tt-retrieval
RAG / context compression: chunking, embedding, HNSW retrieval, <retrievable> tag substitution
-
strays
A TUI for Centralized Management of LLM Agents Running on Machines
-
rho-coding-agent
A fast Rust agent harness with a small footprint and opinionated defaults
-
dynamo-mocker
Mock LLM scheduler and KV manager for testing
-
pizen-lynx-embed
Internal embedding provider for Lynx, backed by FastEmbed
-
tt-provider-gemini
Google Gemini provider adapter — separate API shape (systemInstruction, functionDeclarations, streamGenerateContent)
-
acp-cli
Headless CLI client for the Agent Client Protocol (ACP)
-
caliban-provider-openai
OpenAI schema family for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
mcp-legal
Global legal reference MCP server — case law, legislation, regulations, sanctions screening across US, UK, EU, Africa, Canada, Australia
-
crabcode
Rust AI CLI Coding Agent with a beautiful terminal UI
-
kam-ml
Optional ML inference module for kam variant calling
-
looprs
Concise coding assistant REPL — core library
-
aigw
Umbrella crate re-exporting all AI Gateway provider clients
-
claude-rust-tools
Tool implementations for bash and file operations
-
siumai-provider-openai
OpenAI provider implementation for siumai
-
llmtrace-proxy
Transparent proxy server for LLM API calls
-
saorsa-ai
Unified multi-provider LLM API
-
cortiq-engine
Portable inference runtime for the CMF model format, with no ML framework underneath: runs on CPU, and on GPU (Vulkan / Metal / DX12) with the
gpufeature; tokenizer, chat templates and dynamic per-skill weight overlay -
agentisc-relay-adapter-claude
Claude Code CLI adapter for the agentisc-relay
-
agent-io
SDK for building AI agents with multi-provider LLM support
-
opencode-orchestrator-mcp
MCP server for orchestrator-style agents to spawn and manage OpenCode sessions
-
fabrica-cli
A terminal-based coding agent built in Rust
-
swiftide-langfuse
Composable LLM agents, typed task graphs, and streaming RAG pipelines in Rust
-
llama-engine
Narrow-waist engine trait and core types for llama.rs
-
brainwires-provider
LLM provider implementations (Anthropic, OpenAI Chat + Responses, Google Gemini, Ollama, Bedrock, Vertex AI, local llama.cpp) for the Brainwires Agent Framework. Speech (TTS/STT) providers…
-
cera-wasm
wasm-bindgen browser/Node bindings for the cera inference engine
-
zero-session
Local session history, replay, and wrap-up storage for ZERO
-
openai-agents-core-rs
Core runtime primitives for openai-agents-rust
-
arcan-provider
LLM provider implementations (Anthropic Claude) for the Arcan agent runtime
-
sessfind-common
Shared types and utilities for the sessfind workspace
-
codewandler-flux-flow
Flux-Lang: compile an instruction into a typed, deterministic execution graph — the LLM plans, the runtime runs
-
xchecker-lock
File locking for concurrent xchecker executions
-
yule
Local AI inference runtime — verified models, sandboxed execution, signed audit logs
-
jiandu-memory
Deterministic filesystem memory store with BM25/CJK recall
-
aiguard-adapter-shellhook
Shell-hook adapter for Claude Code, Codex, Gemini, Crush, and Cline
-
perspt-tui
Ratatui-based TUI for Perspt
-
cersei-embeddings
Provider-agnostic text embeddings with a pluggable vector index for the Cersei SDK
-
ffs-symbol
Symbol indexing, bloom filters, and tree-sitter outline extraction for ffs
-
papers-rag
Local vector RAG system for academic papers using LanceDB and FastEmbed
-
potato-type
Toppings for your potatoes
-
aegon-cli
CLI entry point for Aegon — watches Claude Code session files and renders them live
-
xybrid-sdk
Developer-facing API for hybrid cloud-edge AI inference: load/run/stream models with declarative routing
-
thuban_backend_gpu
Compute-shader backend for Thuban built on wgpu
-
spadebox-mcp
Spadebox tools exposed through MCP
-
agent-tool-middleware
Middleware pipeline for LLM agent tool call pre/post processing
-
vw-provider-resolver
Provider and model resolver for VibeWindow
-
langfuse-core
Core types and API client for Langfuse
-
punch-extensions
MCP templates, credential vault, and WASM plugin sandbox for the Punch Agent Combat System
-
nexus-memory-orchestrator
Orchestrator core for Nexus Memory System - session management, event bus, and cross-agent sync
-
cleanpy
Python 3.12+ 最佳实践知识库 - MCP 服务器
-
siumai-bridge
Protocol bridge helpers for siumai gateway and facade integrations
-
apr-qa-runner
Playbook executor for APR model qualification testing
-
mimirs-encoding
Semantic vector encoding for MimirsWell using Candle
-
aeonic-telemetry
Observability layer for Aeonic — tracing, cost accounting, metrics
-
oneai-eval
OneAI Evaluation Framework — structured eval suites, metrics, and runner for agent quality assessment
-
omniference
A multi-protocol inference engine with provider adapters
-
agent-fixes
Session management, code navigation, and orchestration tools for AI coding agents
-
mac-agent-rust
AI-powered macOS automation agent using AppleScript, built with Rig and Ollama
-
awaken-ext-deferred-tools
Deferred tool loading with ToolSearch and probability-based deferral for Awaken
-
byokey-types
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
tt-tokenize
Shared token estimation (tiktoken cl100k + heuristic fallback) used by preview, dispatch, and routing
-
tokn-requests
Composable request pipeline stages, profiles, and eventing for tokn
-
uira-agent
Core agent loop, state machine, streaming, and context management for Uira
-
wordchipper-cli
A command-line interface for the Wordchipper tokenizer library
-
sandbox-agent-agent-credentials
Universal API for automatic coding agents in sandboxes. Supports Claude Code, Codex, OpenCode, and Amp.
-
mirror
unifying multiple LLM backends
-
shore-ledger
Conversation ledger and message persistence layer for the Silvershore daemon
-
roder-ext-parallel-search
Agentic software development tools and SDKs for Roder
-
coro-core
Core library for coro-code - A high-performance AI coding agent
-
ferrum-testkit
Testing utilities for Ferrum LLM inference engine
-
talos-agent
Core orchestration logic and the agent turn loop
-
llmoxide
Provider-agnostic Rust SDK for OpenAI, Anthropic, Gemini, and Ollama (streaming + tools)
-
frankensearch-rerank
Cross-encoder reranking for frankensearch (pure-Rust frankentorch + FastEmbed)
-
lc-embeddings
Embedding model implementations for langchainrust — OpenAI, DeepSeek, Qwen, local/ONNX
-
ras-validation
Validated<T> extractor for tool params
-
agent-doc-element-unknown
Fallback unknown element descriptor for agent-doc
-
memra-kv
KV-cache format policy (q8_0/q5_1/q4_0/fp8 block layouts) for the memra inference engine
-
bamboo-hooks
Lifecycle hook dispatch and external script orchestration for Bamboo
-
autotokenizer
我就只是想要rust能有一個簡單的，從hg上拉下config並製作chat prompt的，也這麼難！要我發明輪子，天啊！
-
tandem-data-boundary
Secure data boundary policy and evidence contract types for Tandem
-
sim-lib-agent-runner-http
HTTP-backed model runners for SIM
-
concerto-core
Pure routing logic, eviction policies, and memory accounting for Concerto. Zero IO dependencies.
-
agentic-forge-mcp
MCP server for AgenticForge blueprint engine
-
onnx_graph
ONNX graph parser and execution engine for deep neural networks
-
mcp-analytics
Governed Analytics MCP Server — metrics, dashboards, funnels, cohorts, anomalies, forecasting, and policy enforcement for AI agents
-
blazen-train
Local fine-tuning (SFT + LoRA) for Blazen
-
paigasus-helikon-providers-bedrock
Amazon Bedrock (Converse API) provider for the Paigasus Helikon AI SDK
-
agent-framework-gemini
Google Gemini chat client for agent-framework-rs
-
ccs-cli
Terminal TUI to select Claude Code providers from cc-switch and launch claude
-
hone-core
Agent loop and orchestration for Hone
-
cersei-agentlang
AgentTemplate — a small functional DSL executed on top of the Cersei runtime
-
agntcy-shadi-sandbox
OS-level sandbox policy and process launching for SHADI
-
roder-dynamic-workflows
Agentic software development tools and SDKs for Roder
-
forge-error
Typed error types for Forge gateway dispatcher traits
-
kalosm-streams
A set of streams for pretrained models in Kalosm
-
flash-rerank-server
HTTP server for Flash-Rerank inference
-
synaptic-confluence
Confluence document loader for Synaptic
-
kapsl-loader
Safetensors weight loader and model config parser for the Kapsl native inference path
-
siumai-protocol-anthropic
Anthropic Messages wire codec for siumai
-
agent-doc-session-registry
Pure session registry mutation policy for agent-doc
-
stynx-code-engine
Agentic query engine with tool-use loop
-
scouter-semver
Scouter semver logic
-
el-engine-candle
RuntimeAcl: InferenceEngine over Candle — real CPU forward (ADR-002)
-
colgrep
Semantic code search powered by ColBERT
-
aof-memory
Pluggable memory backends for agent state management
-
deadband-adapter-openai
Deadband integration adapter for OpenAI Agents SDK
-
llmsdk-openai
OpenAI provider for llmsdk (Rust port of @ai-sdk/openai)
-
loregraph
— memory knowledge-graph for AI coding-agent sessions + repos (single binary). PoC: Claude Code chat ingest + repo crunch + graph + canvas.
-
mcp-ads
Ads MCP Server — a digital advertising platform (campaigns, ad groups, ads/creatives, budgets and bids, audiences, performance metrics with derived CTR/CPC/CPA/ROAS, pacing, and optimization suggestions)…
-
skillsrs
Infinite Skills. Finite Context. - Unified MCP server aggregating tools and skills with 99% token reduction
-
lc-schema
Schema types for langchainrust — Message, MessageType, ImageContent, AudioContent, FileContent
-
nous-api
HTTP API for Nous — eval endpoints, score queries
-
bunsen-onnx-gen
Bunsen utility burn (onnx => rust + burnpack) model generation
-
punch-gorillas
Autonomous agent (Gorilla) system for the Punch Agent Combat System
-
cerememory-evolution
Cerememory evolution engine
-
mcp-messaging
Messaging & Notifications MCP server — push notifications, SMS, webhooks, in-app messaging, message queues, broadcasts
-
synaps-tui
Terminal UI layer — ratatui, crossterm, syntect, tachyonfx
-
onnx-genai-preprocess
Native image and audio preprocessing for ONNX generative AI models
-
cloakpipe-tree
CloakTree: vectorless, reasoning-based document retrieval for CloakPipe's optional RAG tooling
-
team-bot
Telegram bot front-end for teamctl managers
-
praxis-core
Core sandbox policy, workspace enforcement, and command runner for the Praxis tool engine
-
docbert-pylate
late interaction (ColBERT) models, vendored into the docbert workspace
-
awaken-runtime-contract
Runtime-facing contract types, traits, and state model for Awaken
-
dynamo-protocols
Protocol types for OpenAI-compatible inference APIs with inference-serving extensions
-
open-kioku-ingest
Repository indexing pipeline for Open Kioku
-
crewride
A high-performance AI proxy service, supporting multiple AI providers (OpenAI, Anthropic, Gemini) with unified API access and cross-provider forwarding capabilities
-
stakpak-tui
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
cliplama
a local model that does literally nothing :3
-
dejadb-llm
Out-of-box LLM provider backends for DejaDB and Deja Loop: OpenAI-compatible, Anthropic, and Ollama adapters over a small blocking HTTP client, implementing deja_loop::LlmBackend, plus…
-
henri
An agentic AI coding assistant for the terminal
-
algocline-mcp
algocline MCP server adapter — thin tool/handler layer over algocline-app
-
lumen-provider-env
The AI-Native Programming Language
-
thuban_model
Model architectures for Thuban: the dense transformer, vision tower, and shared abstractions
-
rho-tui
Terminal UI (ratatui) for the rho coding agent — port of tau's Textual TUI
-
rumus-serve
High-throughput inference server for RUMUS with continuous batching
-
am-compact
Qwen3 inference engine with Attention Management KV-cache compaction (CUDA + Apple Silicon via MLX)
-
vtcode-battery-pack
Curated battery pack of VT Code crates for building Rust coding agents
-
ceylon-next
A powerful AI agent framework with goal-oriented capabilities, memory management, and tool integration
-
ai-agents-template
Template loading and rendering for AI Agents framework
-
ares-mcp
Model Context Protocol client and server for ARES
-
lc-sessions
Session management for langchainrust
-
ras-llm-vercel
Vercel LLM adapter
-
openclaw-scan
Security scanner for agentic AI framework installations (OpenClaw, Claude Code, and compatible)
-
codewandler-flux-evidence
flux evidence: typed observations at lifecycle phases, an evidence log, and reactions
-
smos-application
SMOS application layer — use cases and port traits
-
llm-content-blocks
Typed fluent builder for Anthropic Messages-API content blocks (text, image, tool_use, tool_result, document). Emits the exact JSON shape the API expects. No SDK dependency.
-
oatmeal
Terminal UI to chat with large language models (LLM) using different model backends, and direct integrations with your favourite editors!
-
life-os
Life Agent OS — install with
cargo install life-osfor thelifeCLI and all subsystem libraries -
shore-daemon
Persistent AI character engine daemon — long-running process behind the Silvershore chat protocol
-
xchecker-runner
Process execution with timeouts and job control
-
a2a-server-rust-lightsail
Minimal Rust A2A server agent
-
tandem-incident-monitor
Incident Monitor domain logic for Tandem
-
llmy-cli
All-in-one LLM utilities
-
offcode
Offline AI coding assistant powered by Ollama
-
adk-ralph
Ralph - Multi-Agent Autonomous Development System powered by ADK-Rust
-
bamboo-llm
LLM provider layer (providers, protocol adapters, streaming, routing, model catalog) for the Bamboo agent framework
-
polyc
Polychrome SDK umbrella: re-exports the core polyc-* libraries (edge adapter, wire types, agent loop, and LLM provider) so a consumer can depend on one crate
-
ferrum-attention
Fused flash attention for Metal and CPU — production-grade, framework-independent
-
awpak-ai
orchestrate AI agents, commands, and URL calls using an execution graph
-
caliban
User-facing binary for the caliban agent harness
-
banshu-ai
Unified LLM API for open-source models and providers (DeepSeek, Z.AI, MiniMax, Moonshot, and any OpenAI/Anthropic-compatible API)
-
crtx-mcp
MCP stdio JSON-RPC 2.0 server for Cortex — tool dispatch, ToolHandler trait, gate wiring (ADR 0045)
-
onnx-runtime-capi
C ABI for the ORT 2.0 runtime: a libonnxruntime.so drop-in surface (skeleton)
-
llm-tool-mcp
Model Context Protocol (MCP) server integration for llm-tool registries
-
caliban-common
Cross-crate plumbing (paths, env-var expansion, atomic writes, globs, tracing targets) for caliban — internal crate for the caliban binary; no API stability, pin exact versions
-
gpu-worker-ort
gpu worker for onnx
-
oracle_omen_policy
Policy language and evaluation engine for oracle.omen
-
edgecrab-cli
Interactive TUI CLI for the EdgeCrab agent
-
elizaos-plugin-local-ai
elizaOS Local AI Plugin - Local LLM inference for text, embeddings, vision, and audio
-
tauri-plugin-llama-crab
Tauri plugin for llama-crab local inference
-
phx
everyone else is rewriting things in rust
-
rwuerstchen
interface for RWuerstchen image generation models models
-
datasphere
Background daemon that distills knowledge from Claude Code sessions into a searchable graph
-
llm-optimizer-api
Core API types and utilities
-
zerobox-snapshot
Filesystem snapshot and restore for zerobox sandboxes
-
openagents
open framework for building, connecting, and running autonomous agents
-
kapsl-engine-api
Core API traits and types for the Kapsl inference engine
-
mem7-datetime
Zero-dependency date/time utilities for the mem7 memory engine
-
mimirs-judge
LFM2.5-based death-loop judge for MimirsWell agentic information flows
-
exa-async
Exa API client for Rust (async)
-
cinch-rs
Opinionated Rust agent framework — library and CLI
-
navi-plugin-broker
Local agentic engine and terminal-first coding agent
-
mi6-tui
TUI library for mi6 - monitoring agentic coding sessions
-
agent-doc-hash
Content hashing helpers for agent-doc
-
awaken-ext-generative-ui
Server-driven UI component plugin (A2UI) for Awaken
-
statsai-pricing
Model pricing helpers for statsai
-
caliban-provider-ollama
Ollama schema family for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
sandbox-agent-agent-management
Universal API for automatic coding agents in sandboxes. Supports Claude Code, Codex, OpenCode, and Amp.
-
leindex-embed
ONNX embedding worker for LeIndex — runs neural embedding and reranking inference in a separate process
-
starpod-agent-sdk
Claude Agent SDK port for Rust - build production AI agents with Claude
-
privacy_http_sdk
Privacy HTTP SDK for Rust
-
tt-inspect-core
Rule engine + tree-sitter harness for the Inspect static analyzer
-
zsfm-tabdpt
TabDPT (Layer6/TabDPT) zero-shot tabular foundation model: GGUF conversion + Burn in-context-learning inference
-
crux-types
Serializable wire-format types for the crux agentic DSL
-
librlm
Recursive Language Models (RLM) - inference-time paradigm for arbitrarily long prompts
-
brainwires-mcp-client
MCP client, transport, and protocol types for the Brainwires Agent Framework
-
candor-cognitive
Heterogeneous LLM inference: Anthropic, OpenAI, DeepSeek, Gemini, LM Studio, Ollama
-
sparrow-providers
Model provider adapters (Brain trait, OpenAI-compat, Anthropic, Ollama, Responses) for Sparrow
-
bitrouter-guardrails
Compiled regex rules and request-check capability
-
kapsl-backend-abi
Stable C ABI shared by Kapsl runtimes and native inference backend packs
-
cortiq-server
OpenAI-compatible HTTP serving for CMF models, built on the cortiq inference engine
-
llmy-tokenizer
All-in-one LLM utilities
-
ryu-downloads
DownloadCenter artifact-fetch primitive for Ryu: one process-wide registry that owns the lifecycle of every artifact Ryu pulls over the network (chat/embedding GGUFs, engine + agent binaries…
-
rkg-lang-swift
Deterministic repository knowledge graphs and context infrastructure for AI-assisted software engineering
-
devsper-core
Distributed AI Swarm Runtime
-
claude-rust-types
Shared domain types for messages, providers, tools, and permissions
-
lynx-embed
Embedding support for Lynx
-
lc-tools
Built-in tools for langchainrust — Calculator, DateTime, URLFetch, etc
-
tensorlake
SDK for interacting with Tensorlake Cloud APIs
-
llm_client
easiest Rust interface for local LLMs
-
agentisc-relay-adapter-noop
Noop (empty) agent adapter for testing and fallback
-
missiond-core
Core library for missiond - PTY management, semantic terminal parsing, and Claude Code orchestration
-
openai-cost
Calculate OpenAI API call cost from a usage block. Cache-aware (cached_input_tokens), supports GPT-5, GPT-4.1, o3, o4 model families, BYO pricing override. No SDK dependency.
-
kalosm-llama
interface for Llama models
-
awaken-ext-mcp
Model Context Protocol (MCP) client plugin for Awaken
-
codei-llm
终端优先的 AI 编程 Agent，用自然语言在本地仓库中读代码、改代码、跑命令、调试问题。
-
onnx-genai-runtime-config
Typed, process-wide runtime environment configuration for onnx-genai
-
agentic-aegis-mcp
MCP server for AgenticAegis streaming validation
-
rag-toolchain
native library designed to empower developers with seamless access to common Gen AI workflows
-
ollama-fit
Inspect local hardware and rank Ollama library models by compatibility
-
candor-tools
12 agent tools: file I/O, search, shell, test, git (sentinel-gated)
-
bamboo-agent
A fully self-contained AI agent backend framework with built-in web services, multi-LLM provider support, and comprehensive tool execution
-
ras-llm-google
Google LLM adapter
-
thndrs-agent
Provider-neutral coding-agent loop and contracts
-
synaptic-huggingface
HuggingFace Inference API integration for Synaptic: Embeddings
-
roder-configure
Agentic software development tools and SDKs for Roder
-
llm-stack-ollama
Ollama local model provider for the llm-stack SDK
-
mcp-commune
Rust client and server for building discoverable Model Context Protocol (MCP) networks
-
ctxgraph-cli
CLI for ctxgraph — local-first context graph engine
-
llama-cpp-sys-v3
Raw FFI bindings for llama.cpp with runtime DLL loading support
-
byokey-auth
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
clester
End-to-end testing tool for clash - simulates Claude Code hook invocations
-
smartcrab
Core library for SmartCrab workflow orchestration engine
-
punch-desktop
Punch Desktop — native desktop wrapper for the Punch Agent OS
-
candor-orchestrator
7-phase LLM-driven agent loop: Observe→Think→Plan→Build→Execute→Verify→Learn
-
motosan-agent-primitives
Core types and abstract traits for the Motosan agent framework
-
kaccy-ai
AI-powered intelligence for Kaccy Protocol - forecasting, optimization, and insights
-
agent-code-lib
Agent engine library: LLM providers, tools, query loop, memory
-
kvbm-kernels
CUDA kernels for KVBM tensor layout transformation and batched memory copies
-
agentic
Support library for building Agentic MCP and Agent2Agent based systems
-
inferi
Cross-platform GPU LLM inference
-
siumai-provider-gemini
Provider-faithful Google Gemini APIs for siumai
-
atomcode-daemon
Open-source terminal AI coding agent
-
wraith-commands
Slash command handling for the WRAITH AI coding agent
-
lumen-provider-mcp
The AI-Native Programming Language
-
wax-bench
Benchmark types and helpers for wax
-
a2a-server-rust
Minimal Rust A2A server agent
-
lynx-common
Shared utilities and types for the Lynx project
-
albert-api
Multi-provider LLM client for Albert CLI — bridges Anthropic, OpenAI, Google Gemini, Ollama, XAI and the Ternlang API with unified streaming and auth
-
zeph-plugins
Plugin packaging, installation, and management for Zeph
-
magi-openai
OpenAI compatible API SDK for Magi AI agents
-
hedos-runtime
The hedos runtime: execution adapters, sidecar supervision, and the memory governor
-
agent-framework-github-copilot
GitHub Copilot chat client for agent-framework-rs
-
llmuxer
Synchronous Rust client for Anthropic, Gemini, OpenAI, and Ollama with a unified trait interface
-
llama-crab-server
HTTP server for llama-crab
-
ccstat-pricing
Pricing fetcher and cost calculator for ccstat
-
robson-connector-openai
Rust async agent orchestrator for automated development workflows
-
notilens
SDK — task lifecycle notifications for AI agents and background jobs
-
open-kioku-semantic
Semantic search orchestration for Open Kioku
-
llm-token-saver-rs
reducing LLM token consumption in production applications
-
tokn-router
Routing, relay, and proxy orchestration across providers for tokn gateway
-
talu
Safe, idiomatic Rust SDK for talu LLM inference
-
pie-client
Pie Client
-
rskit-embedding
Embedding provider abstractions for vector search
-
codei-config
终端优先的 AI 编程 Agent，用自然语言在本地仓库中读代码、改代码、跑命令、调试问题。
-
coding-agent-hooks
Agent-agnostic hook protocol types and adapters for AI coding agents
-
oxios-ouroboros
Ouroboros spec-first protocol for Oxios
-
roder-ext-inference-router
Agentic software development tools and SDKs for Roder
-
synaptic-voyage
Voyage AI embeddings and reranker for Synaptic
-
tandem-automation
Automation model and execution primitives for Tandem
-
thuban_kernel
WGSL kernel registry and dispatch for Thuban
-
vex-router
Intelligent LLM Routing for VEX Protocol
-
voxtral-micro
Voxtral Micro - Minimal text-to-speech with Q4 GGUF quantization
-
team-mcp
MCP server providing the shared agent mailbox for teamctl
-
claude-rust-memory
Session persistence with JSON file storage
-
sbv2_core
Style-Bert-VITSの推論ライブラリ
-
blazen-ggml-sys
Vendored ggml C library, built once and reused by Blazen's whisper-rs-sys / llama-cpp-sys-2 / diffusion-rs-sys consumers via patch.crates-io
-
bar-metrics
Computes BAR Observatory's five-channel effort ledger (tokens, reasoning, latency, drop-rate, repair cost) for Claude Code sessions
-
neurogrim-mcp
MCP client + server integration for NeuroGrim
-
nous-heuristics
Inline heuristic evaluators for Nous — fast (< 2ms), no I/O
-
bamboo-metrics
Live metrics pipeline (bus, collector, worker, aggregation) for the Bamboo agent framework
-
ras-llm-mistral
Mistral LLM adapter
-
llmsdk-provider-utils
HTTP, SSE and config utilities shared by llmsdk providers (Rust port of @ai-sdk/provider-utils)
-
hive-llm
Hive LLM providers (Fireworks / OpenAI-compatible)
-
llm_usages
A CLI tool to track and visualize LLM token usage from Claude Code across projects
-
steer
Command-line interface for Steer coding agent
-
indusagi
— a 100% Rust AI coding-agent framework: LLM gateway, agent runtime, tool/capability kernel, MCP interop, terminal UI, SaaS connectors, multi-agent swarm, and the CLI shell, all in one crate
-
cinch-tui
Generic terminal UI for cinch-rs powered agents
-
langgraph-prebuilt
Prebuilt agents, layouts and tools for LangGraph applications in Rust
-
woollama-engine
The pure-Rust woollama engine (no PyO3): inferencer registry, complete/stream, the recipe<->tool orchestrate loop. Consumed by the woollama-core wheel and the woollamad daemon (woollama-server crate).
-
cerememory-adapter-claude
Cerememory LLM adapter for claude
-
rho-core
Core types and agent loop for the rho AI coding agent
-
loom-core
Core library for LOOM — config, workspace, git, sync, agent, and TUI logic
-
navi-napi
Local agentic engine and terminal-first coding agent
-
claude-pool-mcp
Thin MCP server exposing claude-pool as tools
-
cerememory-store-procedural
Cerememory procedural store implementation
-
scouter-client
Client and server contract for scouter
-
memra-cli
Model onboarding compiler for the memra inference engine
-
mcp-real-estate
Global property intelligence MCP server — transactions, valuations, market data, geocoding across US, UK, EU, Africa, Middle East, Asia-Pacific
-
algocline-nn
algocline neural-network bridge — thin candle wrapper exposing alc.nn to Lua
-
siumai-protocol-gemini
Google Gemini wire codecs for siumai
-
agnt5-sdk-core
High-performance Rust core for AGNT5 SDK
-
oxibrain-ports
Port traits for oxibrain — LLM, embedding, tokenizer, rerank, clock
-
stakpak-mcp-server
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
carter
— a Rust agent CLI with an interactive TUI, multi-provider LLM support, skills, and MCP integration
-
blitzdenk
A minimal multi api auto-context project chat bot as tui
-
gitgraph-core
Core library for GitGraph: git graph parsing, search, actions, and state
-
oneai-agent
OneAI agent paradigms (Plan, ReAct, Reflection, Parallel)
-
mhome-core-api
Shared wire types and service contracts for the mHome core
-
asimov-huggingface
ASIMOV Software Development Kit (SDK) for Rust
-
uira-mcp-client
MCP client runtime for consuming external MCP tools
-
nexus-memory-embeddings
Embedding service for Nexus Memory System using ONNX Runtime
-
bamboo-a2a
Agent-to-Agent (A2A) protocol client for the Bamboo agent framework
-
brainwires-tool-system
Built-in tool implementations for the Brainwires Agent Framework
-
crtx-llm
Claude, Ollama, and replay adapters behind a shared trait
-
argus-difflens
AI code review platform — your coding agent shouldn't grade its own homework
-
dynamo-kv-router
KV Router - Radix tree for LLM KV cache routing
-
tibba-model-token
token billing models for tibba (account, recharge, usage, price)
-
blazen-provider-groq
Generated blazen-llm provider wrapper for Groq (groq)
-
ras-mcp
MCP (Model Context Protocol) server and client
-
tokn-provider-openai
OpenAI and Codex provider integration for tokn
-
sapient-models
Pre-built LLM architecture graph builders for SAPIENT — Llama, Mistral, Phi, Gemma, GPT-2, BERT, Qwen
-
serdes-ai
Type-safe, production-ready AI agent framework for Rust - a full port of pydantic-ai
-
synaptic-condenser
Context condensation strategies for the Synaptic AI agent framework
-
oxibonsai-runtime
Inference runtime, sampling, tokenizer, and server for OxiBonsai
-
jiandu-mcp
Stdio MCP server for the Jiandu shared memory system
-
hop-top-aim
AI model registry client backed by models.dev
-
conduit-ai
Conduit — lightweight AI driver for OpenAI-compatible, Ollama, Anthropic, Gemini (Rust, sync)
-
synaptic-langfuse
Langfuse observability integration for Synaptic
-
synaptic-mcp
MCP (Model Context Protocol) adapters for Synaptic — Stdio, SSE, HTTP transports
-
simple-agents-providers
LLM provider implementations for SimpleAgents
-
agent-stream-kit
Agent Stream Kit
-
sapient-tokenizers
HuggingFace-compatible tokenizers for SAPIENT — BPE, WordPiece, SentencePiece, chat templates
-
wraith-lsp
LSP client integration for the WRAITH AI coding agent
-
thoughts-mcp-tools
MCP tool wrappers for thoughts-tool using agentic-tools framework
-
prime
A CLI tool for interacting with LLMs
-
infernum
CLI - From the depths, intelligence rises
-
kreuzberg-orp
Kreuzberg fork of orp — lightweight framework for ONNX runtime pipelines. Use upstream
orpunless you depend on Kreuzberg. -
synaptic-flashrank
Fast local cross-encoder reranker for Synaptic (BM25-based, zero external dependencies)
-
mc-provider
LLM provider implementations for magic-code (Anthropic, OpenAI, Gemini)
-
mcp-medical
Clinical reference and global health intelligence MCP server — PubMed, WHO GHO, health statistics, medical literature
-
nous-judge
Async LLM-as-judge evaluators for Nous — plan quality, adherence, task completion
-
sevensense-embedding
Embedding bounded context for 7sense bioacoustics - Perch 2.0 ONNX integration
-
toolcap
specifying tool use permissions in agentic applications
-
qex-core
Core library for QEX — semantic code search with BM25, tree-sitter chunking, and optional dense vectors
-
mii-loop
Supervised agentic task loops for local workspaces
-
telepi
Telegram bridge for the Pi coding agent — Rust implementation
-
frankensearch-embed
Embedder implementations for frankensearch (hash, model2vec, fastembed)
-
wikidesk-shared
Shared types and utilities for wikidesk server and client
-
sandbox-agent-universal-agent-schema
Universal API for automatic coding agents in sandboxes. Supports Claude Code, Codex, OpenCode, and Amp.
-
ruvector-onnx-embeddings
ONNX-based embedding generation for RuVector - Reimagined embedding pipeline in pure Rust
-
kapsl-kernels
Compute kernels for matrix operations and attention in Kapsl
-
rkg-lang-fsharp
Deterministic repository knowledge graphs and context infrastructure for AI-assisted software engineering
-
mcp-education
Education LMS MCP server — courses, assignments, grades, students, announcements via Canvas LMS and generic LMS APIs
-
zeph-index
AST-based code indexing and semantic retrieval for Zeph
-
codei-tools
终端优先的 AI 编程 Agent，用自然语言在本地仓库中读代码、改代码、跑命令、调试问题。
-
inferd-engine
Backend trait and adapters for inferd: llamacpp (FFI, runtime accelerator selection), openai-compat, bedrock-invoke, mock. Generation (typed content blocks + JSON-Schema→GBNF grammar) and embeddings.
-
oven-sdk-cohere
Registry-free Cohere v2 Chat adapter for oven-sdk
-
navi-cli
Local agentic engine and terminal-first coding agent
-
unillm-scheduler
Request scheduling with continuous batching for UniLLM
-
symposium-acp-agent
Symposium-enriched ACP agent that wraps downstream agents with enhanced capabilities
-
sandbox-agent-error
Universal API for automatic coding agents in sandboxes. Supports Claude Code, Codex, OpenCode, and Amp.
-
logit-loom-llamacpp
llama.cpp adapter for Logit Loom token-stream mechanics
-
cerememory-archive
CMA (Cerememory Archive) export/import based on SQLite
-
synaptic-postgres
PostgreSQL integration for Synaptic
-
stynx-code-compact
Context compaction and conversation summarization
-
dynamo-data-gen
Shared schemas and primitives for Dynamo data generation (e.g. Mooncake replay traces)
-
life-relay
Remote agent sessions facade for the Life Agent OS — re-exports life-relay-core and life-relay-api
-
synaptic-voice
Voice TTS/STT providers for the Synaptic AI agent framework
-
hanzo-agent
Agent framework with tool calling capabilities for Hanzo AI
-
caliban-provider-google
Google Gemini schema family for the caliban agent harness — internal crate for the caliban binary; no API stability, pin exact versions
-
outis-core
RTEN/ONNX Runtime based captcha recognition engine
-
scouter-observability
Scouter observability logic
-
awaken
Rust AI agent backend: build capabilities once, tune behavior live, serve every client
-
safetensors_explorer
CLI utility to inspect and explore .safetensors and .gguf files
-
kapsl-llm
Large language model inference with GGUF and ONNX backend support for Kapsl
-
potato-state
Holds global state for managing python and rust interop
-
tokn-endpoint-messages
Typed schemas for OpenAI-compatible messages endpoints
-
potato-workflow
Put the potatoes to work
-
synaptic-secrets
Secret management and output masking for the Synaptic AI agent framework
-
ollama2llama
constructing and managing a llama-swap configuration file with Ollama models
-
cerememory-config
Configuration management for Cerememory (TOML + env)
-
agentic-aegis-ffi
FFI bindings for AgenticAegis
-
rag-cli
Local-first RAG CLI for semantic search over your files
-
agent-framework-copilotstudio
Microsoft Copilot Studio agent client
-
onnx-runtime-cpuinfo
CPU hardware detection via pytorch/cpuinfo FFI bindings
-
llama-cpp-sys-4
Low Level Bindings to llama.cpp
-
claude-rust-auth
Authentication with API keys and macOS Keychain OAuth
-
punch-wire
P2P networking protocol for the Punch Agent Combat System
-
anthropic-api-types
Rust type definitions for the Anthropic API (messages, models endpoints)
-
tt-provider-local
Local-LLM provider adapter — Ollama, vLLM, LM Studio (all OpenAI-compatible)
-
candor-sandbox
Dual-engine WASM (wasmtime) + process (bubblewrap) sandbox with circuit breaker
-
homeboy
Headless automation for agentic software engineering workflows
-
wraith-api
AI provider API client for the WRAITH coding agent
-
terraphim_kg_agents
Specialized knowledge graph-based agent implementations
-
candelabra
Desktop-friendly GGUF LLaMA inference wrapper for Candle and Hugging Face Hub
-
open-kioku-tests
Fixture and test selection support for Open Kioku
-
convergio-inference
LLM inference router — model selection, streaming, cost tracking
-
sparkle-mcp
Sparkle AI Collaboration Identity Framework MCP server
-
axonerai
A type-safe, blazing fast agentic AI framework in Rust
-
dynamo-tokenizers
Standalone HuggingFace, tiktoken, fastokens, and Baseten tokenizer implementations for LLM inference serving
-
hone-mcp
MCP client for Hone (Layer 1 external integrations)
-
aigw-openai-compat
Configurable OpenAI-compatible provider for AI Gateway
-
blazen-llm
LLM provider abstraction layer for the Blazen workflow engine
-
agentic-tools-utils
Shared utilities for agentic-tools ecosystem: pagination, http, secrets, cli
-
linear-schema
Cached Linear GraphQL schema for cynic
-
qts
Qwen3 TTS inference (GGUF + GGML); Rust API for host apps and gdext
-
tandem-governance-engine
Premium governance and recursive-authoring policy engine for Tandem
-
stakpak-mcp-proxy
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
llmy-harness
All-in-one LLM utilities
-
candle-pipelines
intuitive pipelines for local LLM inference in Rust, powered by Candle. Inspired by Python's Transformers library.
-
aprender-mcp
Model Context Protocol (MCP) server for aprender — exposes apr CLI as MCP tools
-
autoagents-protocol
Agent Framework for Building Autonomous Agents
-
aiway-model-protocol
The aiway model protocol lib
-
terraphim_kg_orchestration
Knowledge graph-based agent orchestration engine for coordinating multi-agent workflows
-
nemo-relay-cli
Coding-agent gateway CLI for NeMo Relay observability
-
ascend-tools-mcp
MCP server for the Ascend Instance web API
-
lynx-storage
Storage layer for Lynx
-
serdes-ai-providers
Provider abstractions and registry for serdes-ai
-
uira-cli
CLI entry point with session management for Uira
-
mc-config
Configuration loader for magic-code AI coding agent
-
memra-runtime
CUDA runtime scaffolding (context, GEMM checks) for the memra inference engine
-
sapient-io
Model format support: ONNX, GGUF, Safetensors for SAPIENT
-
oxide-agent
Type-safe, high-performance Rust crate for building agentic systems on Ollama
-
plugins-protocol
Newt-Agent provider-plugin JSON-RPC schema + reference client SDK
-
mcp-workflow
Workflow engine MCP server — state machines, approvals, task management, case routing, work orders, change control
-
convergio-mcp
MCP server for Convergio — exposes daemon tools via rmcp SDK
-
synode
A CLI tool for codebase analysis and MCP server for AI-assisted development
-
waycast
A self-hostable observability and analytics platform for LLM applications
-
thuban_backend
Engine-facing backend abstraction shared by the gpu and cpu implementations
-
rho-provider
LLM provider implementations for rho (Anthropic, OpenAI)
-
kernex-providers
AI backend providers for Kernex (Claude Code, Anthropic, OpenAI, Ollama, OpenRouter, Gemini, MCP)
-
concerto-gpu
GPU monitoring abstractions for Concerto: GpuMonitor trait, mock implementation, and optional NVML-backed implementation
-
yus-utils
functions for yus WebAssembly runtime
-
steer-proto
Protobuf definitions for Steer coding agent
-
zeph-experiments
Experiment engine for adaptive agent behavior testing and hyperparameter tuning
-
discord-tools
Discord search tools for agentic-mcp
-
arknet-sdk
Public Rust SDK for building applications on arknet
-
ras-llm-oci
Oci LLM adapter
-
stakpak-gateway
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
infernum-server
HTTP API server for local LLM inference
-
uira-memory
Local semantic memory system for Uira with hybrid search
-
kokoro-cli
A native Rust CLI for Kokoro TTS using XDG standards and agent-first design
-
siumai-extras
Optional utilities for siumai: schema validation, telemetry, and server adapters
-
swiftide-macros
Composable LLM agents, typed task graphs, and streaming RAG pipelines in Rust
-
ferrum-scheduler
Request scheduling for Ferrum LLM inference engine
-
siumai-provider-minimaxi
MiniMaxi provider implementation for siumai
-
dragen
CodeAct-style agent framework using Littrs sandbox
-
blazen-audio-whispercpp
Local speech-to-text backend for Blazen using whisper.cpp
-
orpheus
A blazing fast OpenRouter SDK
-
zerobox-utils-home-dir
Sandbox any command with file, network, and credential controls
-
tandem-graph-core
Shared context graph primitives for Tandem
-
unillm-kv
Hybrid KV cache (RadixAttention + PagedAttention) for UniLLM
-
mlex
Safe, idiomatic Rust runtime for Apple MLX: quantized LLM inference (Qwen, Gemma4, NemotronH, DharaAR, ...) with multi-modal support
-
paddler_messaging
Open-source LLMOps platform for hosting and scaling AI in your own infrastructure
-
byokey-config
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
bamboo-client-core
Shared HTTP/SSE wire types for Bamboo client front-ends (CLI, TUI)
-
roder-api-transcript
Agentic software development tools and SDKs for Roder
-
llm-kit-huggingface
Hugging Face provider for LLM Kit
-
blazen-manager
VRAM-aware model manager with LRU eviction for the Blazen workflow engine
-
ag-harness-cli
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
lha
Long-Horizon Agent command-line package that installs the lha binary
-
harness-tools
Umbrella re-export for harness-* AI agent tools: core, read, write, grep, glob, bash, webfetch, websearch, lsp, skill
-
synaptic-sandbox
Container sandbox for secure code execution — Docker and Apple Container backends
-
vex-llm
LLM provider integrations for VEX
-
synaptic-runnables
LCEL composition: Runnable trait, pipe operator, Lambda, Parallel, Branch, Fallbacks
-
codei-sdk
终端优先的 AI 编程 Agent，用自然语言在本地仓库中读代码、改代码、跑命令、调试问题。
-
oalacea-parry
CLI for Parry - the agentic linter for AI-generated code
-
codewhale-tui
Terminal UI for open-source and open-weight coding models
-
xybrid
Hybrid cloud-edge AI inference SDK — umbrella crate re-exporting xybrid-sdk
-
imp-tui
Terminal UI for the imp coding agent
-
apr-qa-gen
Property-based scenario generator for APR model qualification
-
nenjo-sessions
Shared session contracts for Nenjo runtime, worker, and future session services
-
tandem-plan-compiler
Mission and plan compiler boundary for Tandem
-
bamboo-tools
Tool execution and integrations for the Bamboo agent framework
-
oneiros-mcp
MCP server adapter for oneiros
-
ferrum-interfaces
Core trait contracts for the Ferrum LLM inference engine
-
navi-core
Local agentic engine and terminal-first coding agent
-
tokn-provider-copilot
GitHub Copilot provider integration for tokn
-
edda-notify
Push notification dispatch for edda workspace events
-
scouter-tracing
Helper crate for tracing instrumentation
-
open-kioku-core
Shared data models and schema types for Open Kioku code intelligence
-
rho-engine
rho turn orchestration, provider assembly, tools, and MCP runtime
-
xchecker-tui
Terminal user interface for xchecker
-
ferrum-models
Language model architectures for Ferrum inference
-
deepseek-secrets
Secret storage backends (OS keyring with file fallback) for DeepSeek workspace
-
agentic-forge-ffi
FFI bindings for AgenticForge blueprint engine
-
byokey-daemon
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
concerto-api
HTTP API and orchestrator state machine for Concerto: routes, request lifecycle, cold-start dedup, eviction, and graceful shutdown
-
zsfm-timesfm
TimesFM 2.5 200M (Google) zero-shot time-series forecaster: GGUF conversion + Burn inference
-
agtrace-sdk
Public SDK for building observability tools on top of agtrace
-
cerememory-transport-mcp
MCP (Model Context Protocol) transport binding for CMP
-
cerememory-adapter-common
Shared utilities for Cerememory LLM adapters
-
tandem-runtime
Runtime utilities for Tandem
-
apr-qa-report
Popperian report generator and MQS scoring for APR model qualification
-
synaptic-session
Store-backed session lifecycle management for the Synaptic AI agent framework
-
thuban_gpu
Thin portable compute-shader layer over wgpu for Thuban
-
tandem-tools
Tooling and integrations for the Tandem engine
-
claude-rust-permission
Interactive terminal permission checker
-
ai-agents-cli
Command-line interface for running YAML-defined AI agents
-
tirami-infer
llama.cpp inference engine integration for Tirami
-
nous-middleware
Arcan Middleware implementation for Nous — wires evaluators into the agent loop
-
bamboo-subagent
Sub-agent fleet runtime: project-keyed session store, indices, and Maildir-style mailbox
-
siumai-provider-togetherai
TogetherAI provider implementation for siumai (reranking)
-
bamboo-plugin
Local plugin bundles (manifest, install provenance, installer skeleton) for the Bamboo agent framework
-
greentic-aw-runtime
Enterprise Agentic Worker runtime — Plan-Act-Observe loop, Redis state, tool dispatch via greentic-ext-runtime
-
pictor-runtime
Inference runtime, sampling, tokenizer, and server for Pictor
-
cf-gears-model-registry-sdk
SDK for the Model Registry: API traits, models, and error definitions
-
abu-mcp
MCP protocol implementation
-
nous-lago
Lago persistence bridge for Nous — publishes and subscribes to eval events
-
llm-stack-openai
OpenAI GPT provider for the llm-stack SDK
-
tandem-tui
Terminal user interface for the Tandem engine
-
turul-llm-openai
OpenAI-compatible adapter for the turul-llm LlmClient trait. Targets /chat/completions with response_format = json_schema.
-
awaken-ext-reminder
Rule-based context message injection plugin for Awaken
-
memra-lanes
Serving-lane types, SLO admission policy, and engine-truth step stats for the memra inference engine
-
ao-plugin-agent-claude-code
Claude Code agent plugin for the ao-rs orchestrator
-
navi-providers
Provider facade for NAVI — re-exports the provider implementation crate
-
pakx-agents
pakx agent adapters — Claude Code, Cursor, Codex, Copilot, Windsurf
-
steer-grpc
gRPC server and client implementation for Steer
-
tt-preview
Cost preview engine — projects cost + savings + route suggestions for an LLM request without calling any model
-
tokn-mock-server
Shared test fixtures for tokn-router workspace
-
ras-llm-deepseek
Deepseek LLM adapter
-
roder-config
Agentic software development tools and SDKs for Roder
-
tokn-accounts
Account pooling, affinity, and model route resolution across tokn providers
-
cerememory-index
Cerememory index engine
-
adocs
Agent docs (adocs) - Navigation map for codebases
-
autoagents-toolkit
Agent Framework for Building Autonomous Agents
-
life-vigil
Observability foundation for the Life Agent OS — OpenTelemetry tracing, GenAI semantic conventions, and contract-derived instrumentation
-
dkdc-lm-cli
CLI for dkdc-lm: local LLM inference management
-
siumai-provider-ollama
Ollama provider implementation and protocol standard for siumai
-
tt-provider-together
Together AI provider adapter (OpenAI-compatible)
-
aria-router-agent
Lightweight in-process builtin agent for aria-router
-
rigs
A orchestration framework for rig
-
kyro
A high-performance ML inference engine
-
inferno-ai
Enterprise AI/ML model runner with automatic updates, real-time monitoring, and multi-interface support
-
vestige-engine
Vestige orchestration layer — hybrid search and embed ingest over Store + EmbeddingProvider
-
inferlab-proxy
The built-in serving proxy for Inferlab experiment topologies
-
saorsa-agent
AI coding agent runtime with tool execution
-
zerobox-protocol
Sandbox any command with file, network, and credential controls
-
codei-agent
终端优先的 AI 编程 Agent，用自然语言在本地仓库中读代码、改代码、跑命令、调试问题。
-
xchecker-error-reporter
Human-friendly error reporting for xchecker
-
hone-providers
Model provider abstraction and adapters for Hone
-
potatohead-macro
Macros for Potato
-
codineer-api
AI provider API clients and streaming for Codineer
-
apr-qa-certify
Model certification tools and README synchronization
-
paddler_balancer
Open-source LLMOps platform for hosting and scaling AI in your own infrastructure
-
siumai-provider-azure
Azure OpenAI provider implementation for siumai
-
concerto-backend
Backend process lifecycle for Concerto: BackendManager trait, mock implementation, and a real process-spawning implementation for inference engines
-
mem7-reranker
Reranker trait and providers for mem7
-
cersei-agentrl
AgentRL — self-evolving orchestration (ExecutionGraph, PlannerAgent, dynamic tool registry) for Cersei
-
roder-ext-gemini
Agentic software development tools and SDKs for Roder
-
agent-memory-store
key-value memory store for AI agents — persist facts across turns
-
llmy-agent
All-in-one LLM utilities
-
edda-aggregate
Cross-repo aggregation queries and rollup statistics for Edda
-
kapsl-monitor
Prometheus metrics and monitoring for the Kapsl inference engine
-
tensor-man
A small utility to inspect and validate safetensors and ONNX files
-
open-kioku-embeddings
Embedding provider interfaces for Open Kioku
-
mem7-embedding
Embedding client abstraction for the mem7 memory engine
-
argus-review
AI code review platform — your coding agent shouldn't grade its own homework
-
claude-code-sdk-rs
Rust SDK for Claude Code CLI - Build agents with skills, tools, and orchestration
-
skopos-core
Read, check and write OpenSkopos documents
-
awaken-server
Multi-protocol HTTP server with SSE, mailbox, and protocol adapters for Awaken
-
modelexpress-server
High-performance gRPC server for model serving and management
-
oxibrain-llm-http
HTTP LLM provider adapter for oxibrain (OpenAI-compatible)
-
imp-llm
Standalone multi-provider LLM streaming client
-
ras-llm-cerebras
Cerebras LLM adapter
-
ag-store
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
candor-memory
SurrealDB-backed memory with HNSW vector index and auto-compaction
-
autonomicd
Daemon binary for the Autonomic homeostasis controller
-
warble-vercel
Warble back-end for the OSS harness bundle target (Vercel AI SDK): IR → file-based JSON bundle, emit-only
-
agent-event-bus
Tiny in-process pub/sub for agent loop events. Sync-only Rust mirror of the Python agent-event-bus library: register handlers by event name (or
*for every event), emit, and remove via opaque Subscription handles… -
car-external-agents
Detection of installed agentic CLIs (Claude Code, Codex, Gemini) for the Common Agent Runtime
-
channel
AI coding harnesses
-
deepseek-mcp
MCP server lifecycle and tool proxy compatibility for DeepSeek workspace architecture
-
leone-cuda
CUDA backend and handwritten kernels for Leone
-
scouter-dispatch
Rust Scouter logic
-
siumai-anthropic-compatible
Reusable configured Anthropic Messages-compatible execution engine for siumai
-
zeph-vault
VaultProvider trait and backends (env, age) for Zeph secret management
-
autoagents-llm
Agent Framework for Building Autonomous Agents
-
grimoire-sdk
SDK for building Summoner Grimoire plugins
-
avocode
An AI coding agent with support for 15+ LLM providers, multiple interfaces (TUI/CLI/HTTP), and extensible tools
-
edda-conductor
Multi-phase plan orchestration for Edda
-
deepseek-agent
Model/provider registry and fallback strategy for DeepSeek workspace architecture
-
siumai-provider-anthropic
Anthropic provider implementation for siumai
-
agent-sdk-providers
LLM provider trait, streaming primitives, and first-party provider implementations for the Agent SDK
-
blazen-embed-tract
Pure-Rust ONNX inference backend for Blazen embeddings (tract-onnx)
-
stakpak-ak
Agent knowledge store library
-
stakpak-mcp-config
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
tokn-endpoint-chat-completions
Typed schemas for OpenAI-compatible chat completions endpoints
-
autoagents-guardrails
Agent Framework for Building Autonomous Agents
-
life-praxis
Tool execution engine facade for the Life Agent OS — re-exports praxis-core, praxis-tools, praxis-skills, and praxis-mcp-bridge
-
synaptic-mistral
Mistral AI provider for Synaptic: chat models and embeddings
-
sapient-hub
HuggingFace Hub client for SAPIENT — download any model with one call
-
byokey-provider
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
ras-llm-bedrock
Bedrock LLM adapter
-
siumai-registry
Immutable model routing for siumai
-
deepseek-app-server
Codex-style app-server transport for DeepSeek workspace architecture
-
inferaived
LLM inference engine in Rust using wgpu, with support for Qwen3.5 and MiniCPM5 models
-
byokey-proxy
Bring Your Own Keys — AI subscription-to-API proxy gateway
-
awaken-protocol-a2a
A2A v1.0 protocol wire types and helpers for Awaken
-
memra-validate
Numeric validation helpers (reference comparisons, tolerance gates) for the memra inference engine
-
llmtrace-sdk
Embeddable SDK for Rust applications
-
steer-tools
Tool trait definitions and implementations for Steer
-
tokn-agent-migration
Agent credential and config migration helpers for the tokn-router workspace
-
synaptic-openai
OpenAI integration for Synaptic: ChatModel + Embeddings
-
zerobox-otel
Sandbox any command with file, network, and credential controls
-
lumen-provider-json
The AI-Native Programming Language
-
life-autonomic
Homeostasis controller facade for the Life Agent OS — re-exports autonomic-core, autonomic-controller, and autonomic-api
-
candor-mcp
MCP client (Model Context Protocol) with JSON-RPC 2.0 over stdio/HTTP
-
gba-core
Core execute engine for GBA (Geektime Bootcamp Agent)
-
ferrum-server
OpenAI-compatible HTTP API server for Ferrum inference
-
rusty-logging
Logging for OpsML
-
open-kioku-sentry
Sentry integration helpers for Open Kioku
-
hrdr-llm
OpenAI-compatible chat-completions client for hrdr (streaming + native tool calls)
-
siumai-runtime
Provider-neutral high-level execution runtime for Siumai
-
paddler_client
Open-source LLMOps platform for hosting and scaling AI in your own infrastructure
-
colmena-filter
Output filtering pipeline for Colmena PostToolUse hook
-
candor-graph
Petgraph-based GraphRunner with 17 lifecycle hooks and recovery nodes
-
oxios-web
Web dashboard channel for Oxios
-
zeph-agent-persistence
Agent persistence service for Zeph: history loading, message persistence, tool-pair sanitization, and background extraction enqueueing
-
llama-cpp-v3
Safe and ergonomic Rust wrapper for llama.cpp with dynamic loading
-
inferd
Daemon for the infer AI agent platform
-
ruvector-dag-wasm
Minimal WASM DAG library for browser and embedded systems
-
siumai-provider-xai
Rust-first xAI provider integration for Siumai
-
wax-llm
Command-line LLM inference with Candle, safetensors, GGUF, and Metal support
-
codineer-runtime
Core runtime engine for Codineer: session, config, MCP, prompt, sandbox
-
cerememory-store-common
Shared utilities for Cerememory redb-backed stores
-
potato-agent
Potato brands
-
xchecker-utils
Foundation utilities for xchecker
-
a3s-code-cli
A3S Code CLI - Interactive AI coding agent in the terminal
-
nemo-flow-cli
Coding-agent gateway CLI for NeMo Flow observability
-
roder-protocol
Agentic software development tools and SDKs for Roder
-
claude-rust-provider
Anthropic API provider with SSE streaming
-
polyhook
SDK — write AI coding agent hooks once, run them everywhere
-
ras-llm-langchain
Langchain LLM adapter
-
stynx-code-services
Service layer orchestrating provider calls and tool execution
-
siumai-provider-gateway
Vercel AI Gateway provider implementation for siumai
-
aios-proto
Canonical aios.v1 proto types — generated from core/life/proto/aios/v1/*.proto
-
tandem-wire
Wire-format models for Tandem APIs
-
rho-plugin-sdk
Lightweight SDK for building rho plugins and lifecycle hooks in Rust
-
cersei-memory
Memory trait and backends for the Cersei SDK
-
synaptic-deep
Deep agent harness for Synaptic — planning, filesystem, subagents, skills, memory
-
ruvllm-cli
CLI for RuvLLM model management and inference on Apple Silicon
-
agentic-evolve-ffi
FFI bindings for AgenticEvolve
-
life-lago
Persistence substrate facade for the Life Agent OS — re-exports lago-core, lago-journal, lago-store, lago-fs, lago-api, and lago-policy
-
agent-framework-ollama
Ollama chat client for agent-framework-rs
-
scouter-http
Helper functionality for a common HTTP client
-
roder-ext-runner-sprites
Agentic software development tools and SDKs for Roder
-
siumai-provider-cohere
Cohere embedding, rerank, and audio transcription provider for siumai
-
stakpak-mcp-client
Stakpak: Your DevOps AI Agent. Generate infrastructure code, debug Kubernetes, configure CI/CD, automate deployments, without giving an LLM the keys to production.
-
mc-code-core
Core runtime for magic-code AI coding agent (ReAct loop, agents, memory)
-
lumen-provider-fs
The AI-Native Programming Language
-
synaptic-models
ProviderBackend abstraction and ChatModel wrappers (Retry, RateLimit, StructuredOutput, BoundTools)
-
argus-codelens
AI code review platform — your coding agent shouldn't grade its own homework
-
hone-tui
Ratatui terminal UI for Hone
-
scouter-types
Client and server contract for scouter
-
life-nous
Metacognitive evaluation facade for the Life Agent OS — re-exports nous-core, nous-api, nous-heuristics, and nous-judge
-
claude-rust-engine
Agentic query engine with tool-use loop
-
navi-mcp
Local agentic engine and terminal-first coding agent
-
deepseek-protocol
Codex-style app-server protocol frames for DeepSeek workspace architecture
-
haki
Rust-native coding agent — multi-provider LLM, tools, skills, and a rich TUI
-
thuban_plugin_default
Default model plugin for Thuban: llama, qwen, gemma, and phi architectures
-
tandem-eval
Evaluation harness and regression tooling for Tandem
-
tt-provider-mistral
Mistral provider adapter (OpenAI-compatible)
-
tandem-agent-teams
Agent-team compatibility models and runtime primitives for Tandem
-
ag-session
Agentty is an ADE (Agentic Development Environment) for structured, controllable AI-assisted software development
-
roder-web-search
Agentic software development tools and SDKs for Roder
-
tandem-bug-monitor
Bug Monitor domain logic for Tandem
-
mc-tui
Terminal UI for magic-code AI coding agent (ratatui-based)
-
tokn-provider-deepseek
DeepSeek provider integration for tokn
-
ferrum-engine
Model orchestration engine for Ferrum LLM inference
-
zeph-context
Context budget, manager, compaction strategy, and assembler for the Zeph agent
-
awaken-agent
Compatibility umbrella crate re-exporting awaken
-
ggufy
Unified GGUF wrapper for llama.cpp and Ollama
-
synaptic-callbacks
Callback handlers: RecordingCallback, TracingCallback, CompositeCallback
-
oxillama-runtime
Inference engine — KV cache, sampling, tokenizer bridge
-
potato-provider
Potato brands
-
blazen-llm-candle
Local LLM backend for Blazen using candle inference engine
-
synaptic-memory
Conversation memory: Buffer, Window, Summary, SummaryBuffer, TokenBuffer strategies
-
life-arcan
Agent daemon and runtime facade for the Life Agent OS — re-exports arcan-core, arcand, arcan-harness, arcan-store, arcan-provider, and arcan-aios-adapters
-
ras-llm-groq
Groq LLM adapter
-
synaptic-scheduler
Job scheduling (cron + interval) for the Synaptic AI agent framework
-
open-kioku-tree-sitter
Tree-sitter parser adapters for Open Kioku
-
mezzala
pi coding agent harness in Rust — TUI, multi-provider LLM, WASM extensions
-
autoagents-mistral-rs
Agent Framework for Building Autonomous Agents
-
sapient-telemetry
Structured tracing, metrics, and profiling for SAPIENT (OpenTelemetry-compatible)
-
praxis-mcp-bridge
MCP (Model Context Protocol) server bridge for the Praxis tool engine
-
siumai-provider-elevenlabs
ElevenLabs speech and final-result transcription provider for Siumai
-
roder-ext-perplexity-search
Agentic software development tools and SDKs for Roder
-
synaptic-retrieval
Document retrieval: BM25, MultiQuery, Ensemble, Compression, SelfQuery, ParentDocument
-
agentic-config
Unified configuration system for agentic tools ecosystem
-
siumai-provider-groq
Rust-first Groq Chat, Responses, transcription, and speech provider for Siumai
-
concerto-config
TOML configuration parsing for Concerto
-
kapsl-backends
Pluggable inference backends (ONNX, llama.cpp) for the Kapsl engine
-
scouter
logic