Explore the directory

A Streamlit application, backed by txtai, that answers questions over a person's own data through vector-search RAG and graph-path-traversal RAG, run as a Docker container or a Python virtual environment.

• RAG

An Android app that turns a phone into an OpenAI-compatible LLM server, running models locally through Google's LiteRT-LM runtime on the phone's GPU or CPU and serving them over the local network with no cloud or subscription.

• Inference

An autonomous, AI-driven penetration testing framework that combines large language models with more than 72 professional security tools, aimed at security professionals, bug bounty hunters and enterprise red teams.

• Security

A lightweight MCP server that lets AI assistants such as Cursor and Claude control Spotify playback, search the catalog, and manage playlists through natural-language requests.

• MCP servers

An autonomous, multi-agent AI research loop that reads real literature, forms a hypothesis, runs real machine-learning experiments on a single GPU, reflects on the results, and drafts a peer-reviewed write-up, with guards against reward-hacking.

• Agent frameworks

A Runpod Serverless worker template that deploys an OpenAI-compatible large language model endpoint powered by vLLM, either from a pre-built Docker image pointed at a model or by building an image with the model baked in.

• Inference
• Hosting

A modular, token-efficient AI agent framework, written from scratch in Python, built for local models with a small system prompt and hard, code-enforced limits on what the AI can do, rather than relying on prompt engineering for safety.

• Agent frameworks

A local MCP server that indexes a repository into a symbol graph, giving coding agents a controlled path from compact card-level metadata up to a bounded source-code window, instead of starting from whole files.

• MCP servers
• Agent memory

A complete pipeline that trains a tiny decoder-only Transformer, from a randomly initialized tokenizer and model, on a person's own iMessage history, entirely on a Mac, to learn their phrasing and common replies.

• Models
• Tutorials

RgGen is a Ruby-based code generator for ASIC, IP and FPGA engineers that turns a human-readable register-map specification into SystemVerilog RTL, a UVM register model, C header files and Markdown documentation.

• Languages

R2 Web is a lightweight, browser-only file manager for Cloudflare R2 buckets: it lists, uploads, previews and organizes bucket files directly through R2's S3-compatible API with no backend server involved.

• Hosting

A meta-skill that generates production-ready AI agent skills for Claude Code, OpenClaw, Codex and other platforms from a single target, such as a CLI tool, a REST API, a Python library or a described workflow.

• Agent skills

A macOS menu-bar app that routes the Claude Science desktop client to a third-party model API, such as DeepSeek, Qwen, GLM, Kimi, MiniMax or any OpenAI- or Anthropic-compatible endpoint, and can switch back to the official Claude account at any time.

• LLM gateways
• Desktop utilities

vLLM Kunlun is a community-maintained vLLM hardware plugin that runs large language model inference on Baidu's Kunlun XPU accelerators, supporting more than 20 mainstream and multimodal models.

• Inference

An encrypted, privacy-first on-device AI app for Android that runs chat, vision, retrieval-augmented generation, voice and a sandboxed plugin runtime against local GGUF models, with no Google Play services, telemetry or cloud dependency.

• RAG
• Inference

A Rust CLI and model server for high-performance LLM, VLM, embedding, reranking and audio inference on Apple Silicon and NVIDIA CUDA, executing MLX SafeTensors checkpoints natively with no Python in the request path.

• Inference

A cross-platform desktop app, powered by KoboldCpp, for running large language models and generating images locally on a person's own hardware, with no telemetry, accounts or external servers.

• Inference
• Generative media

A local-first stack that turns a coding agent such as Cursor, Claude Code or Codex into a personal AI assistant for office work, using the browser sessions, desktop apps and tool access a person already has, with no new company platform to approve.

• Agent frameworks

A question-answering dataset and evaluation code for testing retrieval and reasoning across multiple documents in retrieval-augmented generation pipelines, with 2,556 queries whose evidence spans two to four documents each.

• RAG
• AI evals

A curated catalog of open-source and commercial MLSecOps tools and resources for securing machine learning models, MLOps pipelines, LLM applications and AI agents, alongside adversarial ML research and threat-modeling material.

• Security
• Curated lists

The A2A Go SDK is an official Go library for building agentic applications that serve or consume the Agent2Agent protocol, with client and server APIs across gRPC, REST and JSON-RPC transports.

• Agent frameworks

A linguistic linter for Traditional Chinese (zh-TW) that plugs into AI coding assistants over MCP and enforces Taiwan Ministry of Education standards on vocabulary, punctuation and character shapes, catching Mainland Chinese regional drift.

• MCP servers
• Code quality

An open-source batch OCR workbench built around Ollama and GLM-OCR, positioned as a free, local alternative to ABBYY FineReader for digitizing scanned books, documents and forms with a three-panel editor.

• RAG

APEX is a mixed-precision quantization method for Mixture-of-Experts language models that assigns bit-width per tensor type and layer, producing GGUF quantizations for llama.cpp that match or beat full Q8_0 quality at a fraction of the size.

• Inference