Explore the directory

A high-performance inference framework for large language models that targets efficiency, flexibility, and support for multiple hardware vendors, from single GPUs to large clusters.

• Inference

A skill package of thirteen modules that gives a coding agent a structured, evidence-backed workflow for deep reasoning, long tasks, verification, and recovery.

• Agent skills

A small, self-contained web server written in Go with built-in support for Lua and Markdown handlers, HTTP/2 and HTTP/3, and several databases including Redis, PostgreSQL, MySQL, and SQLite.

• Web frameworks
• Backend

A visual, node-based environment for comparing prompts, models, and settings across multiple LLMs and scoring their responses.

• AI evals

A Chinese-language study guide that walks through the internals of the Pi-Agent SDK's architecture and includes a companion agent skill and a worked example for building a data-analysis agent.

• Agent skills
• Tutorials

Apollo CLI is a command-line tool that brings together GraphQL clients and servers: it validates a schema, lints operations for compatibility, and generates static types for client-side type safety.

• CLI tools
• APIs

A collection of sample applications and Jupyter notebooks for getting hands-on with Pinecone vector databases and common AI patterns.

• Tutorials

A tool that lets a coding agent control a real iPhone or Android phone by reading the screen and simulating taps and typing.

• Testing
• Mobile

A reverse proxy for Claude Code that normalizes device fingerprints, telemetry, and identity metadata sent to the Anthropic API so multiple machines look like one canonical client.

• Coding agents
• LLM gateways

An open-source federated search and retrieval-augmented-generation tool that queries an organization's existing apps and data sources live, without copying data into a vector database.

• RAG

A practical guide to Claude Code covering setup, prompt engineering, slash commands, skills, hooks, subagents, agent teams, and MCP servers, from first use to power-user depth.

• Tutorials

An inference engine that links multiple home computers into a cluster over Ethernet to run large language models faster by splitting the work with tensor parallelism.

• Inference

A Chinese-language tutorial and code collection that teaches how to use ChatGPT-style large language model APIs for tasks such as classification, generation, and question answering.

• Tutorials

A self-hosted, offline application that transcribes audio for people who run their own infrastructure and want to keep recordings private rather than use a cloud service.

• Speech

A self-hosted web application that transcribes audio or video to text, translates it, and edits subtitles entirely on your own machine using Whisper models.

• Speech

A general-purpose AI search engine for problems with a known starting point and goal but an unknown path between them, first validated on autonomous penetration testing.

• Security

An educational AI agent microservice designed to be easy to understand and extend, used mainly to teach how AI agents work and to let institutions publish their own agents on the web.

• Agent frameworks
• Tutorials

A lightweight, open-source Kubernetes workspace for multi-cluster operations, observability and access control, with an AI assistant for troubleshooting that respects the user's own RBAC permissions.

• Containers
• Monitoring

A security scanner that discovers every AI agent component installed on a machine, agent harnesses, MCP servers and skills, and scans them for prompt injection and other risks.

• Security

An implementation of a language model architecture that removes matrix multiplication operations, compatible with Hugging Face Transformers, with pretrained checkpoints from 370 million to 2.7 billion parameters.

• Models

A portable data and AI engine, written in Rust, that adds a real-time analytics node to an operational database and serves SQL query, search and LLM inference to applications and agents from one binary.

• Databases
• Data engineering

A six-module open-source course that teaches how to build a production-ready retrieval-augmented-generation AI assistant over a personal knowledge base, using agents, LLMs and LLMOps practice.

• RAG
• Tutorials

An MCP server that connects Ableton Live to Claude and other language models, letting an assistant create tracks, arrange clips and compose music by controlling the DAW directly.

• MCP servers
• Generative media

An open-source .NET framework for building AI agent and chatbot platforms, with a plugin-based pipeline that connects large language models to an existing business application.

• Agent frameworks