Explore the directory

A KV-cache compression library for large language models on Apple Silicon that uses a polar-coordinate quantization scheme to shrink memory use by up to five times while preserving accuracy.

• Inference
• Libraries

An auto-updated leaderboard, refreshed every 15 minutes by a scheduled job, that ranks recently created or updated Claude-skill and AI coding-agent repositories by freshness and momentum rather than all-time popularity.

• Agent skills
• Curated lists

An autonomous multi-agent research orchestrator that runs specialized AI agents for literature mining, data synthesis and hypothesis testing, allocating shared compute as tradeable market credits.

• Agent frameworks
• RAG

A research implementation of retrieval-augmented generation for relation extraction, providing an end-to-end pipeline that runs with and without retrieval across several standard relation-extraction datasets.

• RAG
• Papers

An MCP server that lets an AI agent operate a Proxmox virtualization cluster by planning every risky action first, taking a snapshot where possible, and writing a signed receipt of what changed.

• MCP servers
• Hosting

A production-style document assistant that lets a user upload PDFs such as financial reports or legal contracts and chat with a retrieval-augmented model that cites the exact source page for every answer.

• RAG

A minimal Compose Multiplatform sample showing how to call on-device generative AI from shared Kotlin code, using Apple's Foundation Models framework on iOS and Google's ML Kit Prompt API on Android.

• Inference
• Mobile

An overnight execution layer that takes queued GitHub issues or written PRDs, runs Claude CLI or Codex inside isolated git worktrees while the developer is offline, and opens pull requests to review in the morning.

• Coding agents
• Git

An open source manager, written in Rust, for MCP servers and agent skills across every AI client on a machine, giving one dashboard to install, back up and roll back configuration.

• MCP servers
• Agent skills

An MCP server that connects an AI client such as Claude Desktop or Cursor to a running ZenML instance, giving it read access to ML and LLM pipeline definitions, runs and deployments.

• MCP servers
• Data engineering

Safe Rust bindings to llama.cpp that track the upstream project closely, exposing a high-level API alongside raw bindgen bindings, with example programs for chat, embeddings and an OpenAI-compatible server.

• Inference
• Libraries

A Streamlit web interface for connecting to one or more MCP servers and chatting with them through a chosen language model provider, supporting OpenAI, Anthropic, Google and Ollama.

• MCP servers

A defense-in-depth security toolkit for LLM agents that tracks where every piece of text came from, scans it for prompt injection, applies policy before a tool runs, and scrubs outbound secrets.

• AI evals
• Security

An unofficial, paper-aligned implementation of DeepSeek's Engram architecture that injects a large, sparsely retrieved conditional memory into a transformer model without increasing inference compute.

• Models
• Papers

A real-time .NET reverse proxy and dashboard that intercepts every API request and response between Claude Code and the Anthropic API, so a developer can inspect traffic and MCP tool calls.

• Debugging
• Monitoring

A collection of 36 finished, cloneable mobile app templates that run their AI features entirely on the phone's own hardware, spanning chat, translation, vision, health, voice and forecasting.

• Mobile
• Curated lists

A curated, fully automated index of more than 400 Korean-language AI coding-agent skills, agents, rules and hooks for tools such as Claude Code, Gemini CLI, Codex and Cursor.

• Agent skills
• Curated lists

An autonomous task orchestrator for long-running projects that decomposes a written goal into subtasks, routes them to isolated worker agents, and verifies results against acceptance criteria.

• Coding agents
• Agent frameworks

A local-first memory layer that sits underneath an AI agent's built-in memory, recording which strategies worked or failed on past tasks so future runs can reuse what succeeded.

• Agent memory

A local, cross-tool usage dashboard for AI coding assistants that reads the logs and local databases already kept by Claude Code, Cursor, Kiro, Copilot and Codex to show cost and workflow patterns.

• Coding agents
• Monitoring

A measured serving recipe and benchmark comparing two 4-bit quantized checkpoints of a 27-billion-parameter Qwen model against an FP8 baseline on a single NVIDIA DGX Spark, showing where quantization helps and where it stops mattering.

• Inference
• AI evals

A single-binary Jira command-line client written in Rust, with a mouse-driven terminal interface for browsing, editing and bulk-managing issues, plus a bundled MCP server that exposes Jira as tools for editors and AI agents.

• MCP servers
• CLI tools

The official Java client library for the Pinecone vector database, letting JVM applications create and query Pinecone indexes over a shared HTTP client.

• Databases

A Rust database that speaks the PostgreSQL wire protocol and adds vector and graph access paths on top of the same catalog and on-disk storage engine. It is alpha software, not for production data.

• Databases