Explore the directory

A research framework that improves the faithfulness of retrieval-augmented generation by identifying and suppressing the feed-forward network layers most responsible for a model ignoring retrieved context.

• RAG

A self-hosted MCP server that gives Claude, Cursor, Codex and other Model Context Protocol clients shared, persistent memory with vector, keyword and hybrid search.

• MCP servers
• Agent memory

A marketplace of Claude Code plugins covering the software development lifecycle, bundling skills, commands, agents and workflows for frontend, backend and other domains.

• Agent skills

A self-hosted private AI infrastructure and process-automation platform for Kubernetes, combining local model serving, a chat interface and workflow automation on the operator's own hardware.

• Containers
• Hosting

A self-hosted OSINT investigation platform that turns uploaded documents into an entity graph, then runs an autonomous agent that pivots on open-web infrastructure to extend the investigation.

• Security

An autonomous penetration-testing platform where an LLM agent plans and runs reconnaissance, active testing and vulnerability verification against an authorized target, then writes a report.

• Security

A software development kit for building an AI copilot into a product, with React components for the frontend and a multi-provider LLM SDK for the backend.

• Libraries
• UI kits

An open-source, local runtime for debugging AI agent workflows built as DAG pipelines, with tracing, replay and artifact lineage rather than a framework for building agents.

• AI evals
• Debugging

A course and code repository that teaches how to build autonomous AI agents from scratch in Python using ReAct prompting, without relying on third-party agent libraries.

• Tutorials

A JupyterLab, Notebook 7 and JupyterLite extension that adds AI code completions and a chat panel inside the notebook, configurable with different AI providers.

• Editors

A TypeScript agent framework, rewritten around an entity-component-system architecture, that gives autonomous agent workflows persistent SQLite-backed memory with semantic search, a choice of cloud or local reasoning models, and containerized execution.

• Agent frameworks
• Agent memory

An Open WebUI tool that ports the Superpowers agentic development workflow of brainstorming, spec writing, planning and execution to any local or remote OpenAI-compatible model.

• Coding agents
• Agent skills

A portable local AI environment that runs offline from a USB drive or hard disk, providing text chat, image generation and text-to-speech without installing anything on the host machine.

• Inference
• Generative media

A provider plugin that adds Ollama Cloud as a model source for the Pi coding agent, with dynamic model discovery and web search and fetch tools through an API key.

• Coding agents

An MCP server that connects AI coding agents to a running Neovim session over its native msgpack-RPC socket, without needing a Neovim plugin.

• MCP servers
• Editors

A local AI orchestration platform for running large language models offline, with a FastAPI backend and a SvelteKit interface for chat, document work and task automation.

• Inference

A benchmark site that runs the same prompt through different combinations of model, agent harness and thinking effort, then shows the resulting runnable artifacts side by side for comparison.

• AI evals

A deterministic transformer-shaped virtual machine that compiles a small assembly language into a runtime, executes it with no randomness, and proves the resulting computation with a transparent STARK proof system.

• Security
• Languages

A multi-framework neural-network compiler that lowers PyTorch, JAX, Flax and Keras models into portable GGUF binaries and standalone, human-readable C++ code with CPU and CUDA GPU execution.

• Models
• Inference

A Claude Code skill that delegates implementation-heavy coding work, such as multi-file refactors and boilerplate generation, to the Codex CLI as an execution specialist, while keeping planning, review and final acceptance inside Claude.

• Coding agents
• Agent skills

A Claude Code plugin that gives the agent persistent memory, a name and personality across sessions, and messaging through WhatsApp, Telegram, Discord, iMessage and Slack, running inside Claude Code's own plugin system rather than a separate runtime.

• Agent skills
• Agent memory

A fork of an MCP bridge for Cheat Engine that replaces its Windows named-pipe transport with a native TCP connection, letting AI tools such as Cursor, Copilot and Claude drive local or remote Cheat Engine sessions for reverse engineering, pointer scanning and memory analysis.

• MCP servers
• Debugging

A codebase-understanding tool for AI agents, distributed as an MCP server and command-line tool, that turns a repository into structured static-analysis data, such as a tech-stack summary, symbol outlines and a dependency graph, for a language model to interpret itself.

• MCP servers
• Agent memory

A video-based, project-driven course bridging machine-learning theory and production AI-engineering practice, covering agentic systems, retrieval-augmented generation, orchestration, and fine-tuning large language models.

• Tutorials