Explore the directory

A local-first MCP tool and dashboard for investigating how Codex CLI work consumes tokens, time, tools, allowance, credits and cost, and for spotting caching and thread patterns.

• MCP servers
• Monitoring

A collection of Claude Code and Codex skills, including a load-bearing architecture design and governance skill and a risk-scaled coding-execution protocol with evidence guardrails.

• Agent skills

A sequence of runnable notebooks that teaches time-series anomaly detection from simple z-scores up to modern deep learning models, with honest evaluation and real-world case studies.

• Tutorials

A collection of Java MCP servers built with the Quarkus MCP server framework, covering JDBC databases, JVM inspection, the filesystem, JavaFX drawing, Kubernetes, containers and Wolfram Alpha search.

• MCP servers

A tool that lets AI coding agents save engineering intent and decisions to Git alongside the code, so a later agent can read why a path was tried and abandoned, catch a conflict with another agent's in-progress work, or see the constraints behind risky code before editing it.

• Agent memory
• Git

A Kubernetes operator for self-hosted LLM inference across a mixed fleet of NVIDIA, AMD and Apple Silicon hardware, turning model deployment into a two-line YAML definition with GPU scheduling, caching and OpenAI-compatible endpoints.

• Inference
• Containers

The companion code for a Korean-language book on building practical LLM applications, covering transformer architecture, model training and serving, embeddings and retrieval-augmented generation with LlamaIndex.

• Tutorials
• Books

An Android workbench for studying on-device AI in real scenarios such as online TV, combining llama.cpp, whisper.cpp and a customized FFmpeg to run offline LLMs and real-time transcription while watching or recording video.

• Inference
• Mobile

An MCP server that connects Claude Code, Cursor or Windsurf directly to a running Unity or Godot editor, with hundreds of granular tools, project-aware context, script semantic search and gameplay scaffolders.

• MCP servers

The official implementation of a CVPR 2026 paper on language-aware RGB-thermal object tracking, combining a multi-modal transformer encoder with retrieval-augmented context reasoning.

• Papers

A local-first skill management platform for AI coding agents, combining SQLite and Git persistence, hybrid semantic search, bandit-optimized suggestions and MCP integration to turn operational knowledge into reusable, searchable artifacts.

• Agent skills

A pure-Python library that decides what an LLM actually sees from retrieved documents: retrieval, re-ranking, memory decay and token-budget enforcement in one pipeline.

• Agent memory
• RAG

A documented top-ten list of security risks specific to autonomous AI agent systems, such as authorization hijacking, memory manipulation and multi-agent orchestration exploitation, with mitigation strategies for each.

• Security
• Books

A Harbor-compatible framework for evaluating coding agents such as Claude Code, Codex and Cursor CLI in sandboxed Docker or Modal environments, with support for installing agents in air-gapped tasks and a trajectory viewer.

• AI evals
• Testing

A sample serverless AI agent that uses LangChain.js and the Model Context Protocol to browse a menu, place an order and track status against a burger-ordering API, deployed on Azure Static Web Apps and Functions.

• Agent frameworks
• Tutorials

An MCP server that gives an AI assistant visibility into a 1C:Enterprise configuration's metadata so it can generate accurate BSL code, working with local models or cloud AI clients.

• MCP servers

A Kotlin Multiplatform library for running large language models, speech-to-text and image generation fully on-device across Android, iOS, desktop and WASM, built on llama.cpp, whisper.cpp and stable-diffusion.cpp.

• Inference
• Libraries

A manager dashboard that unifies Claude, Codex, Cursor, Grok, Copilot and other AI coding CLIs you already have subscriptions for into one assistant, with a shared memory, a Kanban board and a Markdown workspace agents can read.

• Coding agents

A macOS menu bar app that quietly logs what you read and work on all day into local Markdown files, then turns that record into a cited knowledge base and daily summary for an AI assistant.

• Agent memory
• Desktop utilities

A native AI framework for Delphi and Free Pascal/Lazarus providing direct provider-specific components for OpenAI, Claude, Gemini and Ollama, plus RAG pipelines, autonomous agents, MCP servers and clients, and native chat UI components.

• Agent frameworks
• Libraries

A protocol-agnostic TypeScript runtime that turns typed functions into services AI agents can discover, call, pay for, stream or run as tasks, supporting crypto and fiat payment rails such as AP2, A2A, x402 and ERC-8004.

• Agent frameworks

A self-hosted agentic AI operating system with a voice-controlled dashboard and a knowledge-graph second brain, letting a developer swap between Claude Code, ChatGPT/Codex, Grok Build, Antigravity CLI or API providers as its backing model.

• Agent frameworks

A self-hosted, open-source financial data MCP server that scrapes and serves SEC filings, XBRL financials, 13F holdings, insider and congressional trades and other market data to AI agents over 62 tools.

• MCP servers
• APIs

An open-source build system that turns written specifications and architecture into working code, having an LLM generate each step under tight constraints and rebuilding only what changed.

• Agent frameworks
• CLI tools