Explore the directory

A plugin for the herdr terminal tool that opens an agent's changes in the hunk diff viewer, collects inline review comments, and sends them back to the agent that made the change.

• Git
• Code quality

An autonomous system that pairs coverage-guided fuzzing with LLM agents to find, verify and patch vulnerabilities in open-source projects, built on the OSS-Fuzz toolchain.

• Testing
• Security

A self-hosted semantic search and memory layer for AI coding assistants that indexes a codebase with local embeddings, aiming to cut context tokens by around 98 percent, installable via Docker or from source and integrated through an MCP client.

• Agent memory
• RAG

A cross-platform desktop app for building, testing and validating MCP tools in a controlled local sandbox before exposing them to an agent, backed by a built-in MCP server, an agentic chat interface with RAG and human-in-the-loop approval, and detailed observability dashboards.

• MCP servers
• Testing

A set of 211 free visual flashcards, one concept per card with a diagram and explanation, covering LLM topics from tokenization and attention through RAG, inference and deployment, organised into guided study paths.

• Tutorials

A repo-native agent that turns private office documents such as decks, spreadsheets, PDFs and emails into a local, evidence-first knowledge base, preserving structure like slide layout, table references and formatting so an AI agent's answers stay traceable to a source file.

• Agent memory
• RAG

A full-stack multi-agent chat application built with React, TypeScript and LangGraph.js, offering a deep researcher, a chat assistant, a math solver and an MCP tool-calling agent behind a gateway with streaming, image input and human-in-the-loop confirmation.

• Agent frameworks
• MCP servers

An automatically generated, LLM-curated and continuously updated directory of extensions, themes, videos and articles for the Pi coding agent ecosystem, indexing thousands of resources through a daily discover-filter-classify-render pipeline.

• Coding agents
• Curated lists

A persistent, multi-role runtime for long-horizon AI research and engineering, splitting planning, execution and read-only review across Manager, Planner, Engineer and Reviewer roles so work can pause, be checked and resume for hours unattended.

• Agent frameworks

A model-agnostic, dependency-free scaffold of Markdown personas and rules for coding agents, where a Maestro orchestrator breaks a plain-language request into sub-agent work handled by an Architect, Coder, Reviewer and Contextualizer, each able to run on a different model.

• Agent skills

A Jax library for quantization-aware training and post-training quantization of Flax models, targeting CPUs, GPUs and TPUs as well as on-device LiteRT deployment.

• Models

A Claude Code skill that downloads a video from YouTube, Instagram, X, Vimeo or almost any yt-dlp-supported site, or a local file, extracts frames and transcribes the audio locally, so Claude can read the frames as images and answer questions about the video.

• Agent skills

A TypeScript video-editing pipeline that takes raw footage and a creative brief and runs it through a coordinated group of agents for shot selection, trimming, rendering and automated quality review, with an optional web studio interface.

• Generative media

A knowledge-graph backed platform that lets multiple AI coding agents share code understanding, plans and decisions through a central Neo4j-based knowledge base, with semantic search and MCP tool integration for Claude Code and other clients.

• Agent memory
• Databases

A self-hostable, open-source AI coding agent that inspects a repository, plans multi-step work, edits files, runs commands and opens pull requests from a browser-based interface, with a choice of hosted or local language models.

• Coding agents

A Model Context Protocol server that gives Claude Code and other AI coding assistants persistent memory across sessions, storing project history, decisions and progress in a local, SQLite-backed store shared across parallel sessions.

• MCP servers
• Agent memory

An MCP server written in Delphi that connects Claude Code, Codex and other MCP clients to Delphi projects, with an extensible RTTI-based tool system, HTTP and stdio transport, and built-in security checks.

• MCP servers

A macOS app that runs open-source large language models locally through llama.cpp, offering both an OpenAI-compatible API for developers and a built-in chat UI for non-technical users.

• Inference

A cross-platform quantization engine for PyTorch models that reduces memory use and increases inference and training speed on GPUs, NPUs, CPUs and ARM devices.

• Models
• Inference

A local-first AI voice studio that clones voices, synthesizes speech through seven TTS engines, provides system-wide dictation, and connects MCP agents to custom voices.

• Speech

An open-source, generic MCP client for testing and evaluating MCP servers and agents from an interactive command-line or web chat interface.

• MCP servers
• Testing

A TypeScript framework for loop engineering that runs an agent toward a stated goal in rounds, with a separate skeptical, read-only verify agent deciding when the goal is actually met.

• Agent frameworks
• AI evals

A self-hostable AI gateway written in Go that routes AI coding tools to one local endpoint, with automatic provider failover, semantic caching, budget limits and guardrails.

• LLM gateways

A 2-bit complex-valued large language model in which every weight is one of four complex values, achieving efficient compression with minimal accuracy loss compared with real-valued 2-bit models.

• Models