Explore the directory

A Claude Code plugin that sends session telemetry, including prompts, tool calls, model generations, token usage and cost, to Langfuse for tracing and observability.

• AI evals

A desktop control plane for discovering, converting, running, chatting with and benchmarking open-weight language and image models entirely on your own machine.

• Inference

A package manager and catalogue of installable skills, subagents and hooks that structure how an AI coding agent moves from an idea through implementation, review and release.

• Agent skills

A hosted API that lets AI agents join a shared chat room and talk to each other over plain HTTP requests, with no signup or SDK required.

• Agent frameworks

A Scala 3 toolchain that compiles straight to a native executable through Scala Native, with a self-hosted compiler, linker and build tool that require no JVM at all.

• Languages

A collection of deployment recipes and practical guides for running the Ling model family locally with SGLang, vLLM, llama.cpp, Ollama and MLX across consumer and workstation hardware.

• Inference
• Tutorials

A dependency-free C and SIMD int8 runtime that runs the FastEnhancer-Medium speech enhancement model in real time, without an inference framework or heap allocation after startup.

• Speech
• Libraries

A command-line tool that parses Common Information Model (CIM) RDF schemas and generates language-specific data model code, with a Go rewrite of the original generator.

• CLI tools
• Libraries

A compression library for embedding indexes and LLM KV caches that optimizes for the metric a downstream consumer actually uses, such as retrieval recall or generation quality, instead of reconstruction error alone.

• Inference
• Databases

An integration that lets the txtai semantic search library use Qdrant as its vector storage backend, including cluster deployments for datasets too large for one machine.

• Databases

A library of 43 KV-cache compression methods for MLX models on Apple Silicon, letting a locally run model use up to 16 times less cache memory with a three-line API change.

• Inference

A self-contained MCP server for undetectable Chrome browser automation, sharing one backend across many concurrent AI agent sessions with named, persistent browser profiles.

• MCP servers
• Browser automation

Enforces the rules already written in a project's CLAUDE.md, AGENTS.md or .cursorrules files, blocking AI coding agents from crossing them using semantic conflict detection rather than keyword matching.

• Agent skills
• AI evals

Forty self-contained tutorials on production retrieval-augmented generation techniques, from chunking strategies to agentic and graph-based retrieval, runnable on any OpenAI-compatible provider.

• RAG
• Tutorials

An agent-native database that stores memory, state, events, artifacts and relations as first-class objects, returning structured evidence instead of plain top-k text matches.

• Agent memory
• Databases

An agent skill that pentests your own running application: it finds a vulnerability, proves it with a real HTTP request, patches it, then re-verifies the fix.

• Agent skills
• Security

A collection of step-by-step tutorials on building and integrating agentic AI systems, published under IBM's Open Innovation Community initiative.

• Tutorials

A multi-interface music analysis and generation toolkit built on the music21 library, offering the same 13 analysis tools through an MCP server, an HTTP API, CLI tools and a Python library.

• MCP servers
• Libraries

A unified KV-cache compression toolkit for LLM inference offering twelve Python-native compression methods behind one API, with guarded add-on composition, multi-GPU deployment planning and launch-command generation for llama.cpp and vLLM.

• Inference

A Python toolkit and interactive Streamlit playground for talking to remote Model Context Protocol servers over Streamable HTTP or SSE, with LLM-powered tool selection across several providers.

• MCP servers

A set of Model Context Protocol servers, in Go, Java, Python and Rust, that expose the Apache Fineract core-banking backend's loan, savings and back-office operations to AI agents.

• MCP servers

A reinforcement-learning training framework that composes an entire multi-agent system at each step rather than building it incrementally, paired with MASBench, a controlled benchmark for when multi-agent systems beat single-agent ones.

• Agent frameworks
• AI evals

A self-hosted text-to-speech engine with 28 voices in ten languages and voice cloning from about ten seconds of audio, with native SDKs for Python, Swift, Go, Rust and TypeScript, running offline after the first model download.

• Speech

A specification-driven task-management framework for AI coding agents that tracks task dependencies as a DAG, enforces quality gates called Constitution rules before a task can complete, and supports pausing, cancelling and recalling tasks.

• Agent frameworks