Code agents

AI Learning: A comprehensive repository for Artificial Intelligence and Machine Learning resources, primarily using Jupyter Notebooks and Python. Explore tutorials, projects, and guides covering foundational to advanced concepts in AI, ML, DL and Gen/Agentic Ai.

Scalable Agentic RAG system using Pydantic AI, FastAPI & pgvector. Modular, production-ready foundation for document-based AI apps

Refactored fork of the sourcemap-leaked Claude Code CLI

The coding agent you can walk away from. Text it a voice note from your phone and it builds while you are out; a second AI reviews every job before you see it. Desktop app for Mac and Windows, on the AI plan you already pay for.

A Model Context Protocol (MCP) server written in Go that empowers AI agents and Large Language Models (LLMs) to seamlessly interact with Kanboard. It transforms natural language commands into Kanboard API calls, enabling intelligent automation of project, task, and user management, streamlining workflows, and enhancing productivity.

AI-Suite - n8n, OpenClaw, Open WebUI, OpenCode, Llama.cpp/Ollama, Flowise, Langfuse, MCP Gateway and more!

Holmes is an interactive, text-based crime investigation game powered by a large language model (LLM). With each replay, the game offers a fresh narrative, ensuring a unique experience for players every time.

Run many Codex & Claude agents in parallel without them overwriting each other. Isolated worktrees, file locks, PR-only merges. Auto-wires Oh My Codex, Oh My Claude, OpenSpec, and Caveman in every worktree.

Open-source AI design studio. Multi-variant canvas, inspect & edit in code. Powered by Claude / GPT. 本地优先的 AI 设计工作台。

MCP server for controlling Miro whiteboards with AI assistants

Delegate complex tasks to Manus AI - web research, report generation, code building, data scraping. Task templates, monitoring, cost tracking. OpenClaw skill.

Open-Source Framework for AI-Driven Quality Assurance. No-Code AI Agents for QA Testers. Automate Web, API, Database, and Mobile testing with natural language.

Ultimate Telegram Bot API MCP server — full v9.6 coverage, meta-mode, rate limiting, circuit breaker. TypeScript.

🍞 Toast notifications from AI coding agents on your macOS menu bar, with tmux pane switching

Self-hosted MCP memory server for personal use and teams

RamiBot v3.8.0 is a local-first AI security operations platform integrating multi-LLM support, a dynamic red/blue team skill pipeline, MCP tool orchestration, Docker terminal access, Tor proxy management, and an auto-integrated Kali-based tool server (rami-kali) for controlled, extensible offensive and defensive workflows

Extract a person into an AI agent — through an 8-week therapist-style interview, not a form. The open-source alternative to $4K AI clone courses.

A modal text editor built for the age of AI agents — Neovim-style editing, Emacs-style runtime flexibility, powered by Elixir/BEAM and Zig

High-performance LLM Gateway built in Go - OpenAI compatible proxy with multi-provider support, adaptive routing, and enterprise features

Aqueduct is a simple yet fully-featured AI gateway built on top of the LiteLLM Router SDK using Django.

Fully automated open-source trading bot for BIST (Borsa Istanbul), a Turkish stock exchange, including both technical indicators and AI news analysis. Implemented with NodeJS.

🇬🇧🏦 MCP server for UK Companies House API - Search companies, retrieve detailed information, filing history, officers, and charges data through the Model Context Protocol

122 AI Agent Skills for full-lifecycle product development (PM × UI × Backend) | 覆盖产品全生命周期的122个AI Agent技能，从探索发现到增长运营，覆盖产品×设计×后端三大领域

See an app. Ship an app. — Turn a screen recording into a production-ready spec.md for Claude Code.

MCP server for .NET debugging — 34 tools for breakpoints, inspection, exception autopsy, code analysis, and more. Built on ICorDebug.

Black-box web penetration testing automation framework for AI Agents

Claude Code plugin：去除文章中的 AI 寫作痕跡，使文字更自然、更有人味。以台灣正體中文為主。

AI harness for and with Eclipse IDE with the focus on offline LLM with llama.cpp. Build with LangChain4J. Supports Ollama, LM Studio, llama.cpp, unsloth or cloud providers like with OpenA API, Anthropic Claude, Google Gemini, Mistral, GitHub Copilot. Contributions are highly welcome!

Run multiple Claude Code, Codex and Gemini subscriptions side by side — shared skills/agents/memory, isolated logins, live 5h/7d limits in one view

Terminal UI toolkit inspired by Claude Code — 25+ React components for building rich CLIs, AI agents, and dashboards

Checks AI-generated code changes before merge: scope, validation, risk, review evidence, and optional Verity receipts.

Expose OpenCode through A2A with an inbound server surface and embedded outbound client.

Manage AWS, Hetzner, OVH, and custom servers from one TUI — with a built-in AI assistant and MCP server

Your personal AI team that gets things done. Shared memory. Any model. Local or cloud. Run. Rest. Dream. 🐨

6 specialized AI agents collaborating via A2A protocol to power e-commerce — product discovery, orders, pricing, reviews, inventory & support. Dual-stack (Python + .NET/C#) with generative UI. Built with Microsoft Agent Framework, FastAPI, Next.js, and PostgreSQL + pgvector.

Professional Agent Skill for building, auditing, testing, and releasing modern WordPress plugins with Codex, Cursor, and Claude Code.

Automate a WeChat group's daily work-report roll-call — start, chase non-submitters, and LLM-summarize for the manager. wxauto + FastAPI + hot-reloadable rules.

Climb from LLM API to RL Agent with Klara

The AI agent memory layer you can audit — local-first memory governance for AI agents: citations, trust policies, trace receipts, rollback. SQLite, sidecar-first, OpenClaw plugin.

MCTS (Model Context Threat Scanner) is a local-first security scanner for MCP servers -- static and live tool discovery, multiple analyzers, auditable risk scores, and JSON, SARIF, and HTML output. For authors and platform teams; CI-ready, no cloud API.

A browser extension that allows you to highlight, tag, annotate, and export the best parts of your favorite YouTube videos and summarize these snips with the power of AI. Winner of Boost Hacks

Turn your AI assistant into a senior web developer. Production-ready guides for SEO, performance, and agent-readiness - for Claude Code, Cursor, Copilot, and Windsurf.

TriLane: autonomous AI pentester and multi-agent security auditor for authorized gray-box testing of web apps and APIs. Built with Rust and Tauri.

Agent harness research ebook — model-agnostic architecture specification (June 2026)

Offline agent skill bundling specification.website (128 topics) for Claude Code and other SKILL.md-compatible agents. Spec content CC BY 4.0 by Joost de Valk.

General-purpose agent superteam for Claude code - contract-gated verification loops with warm-start local wiki for self-evolving, and structural prevention of premature exits.

Make your repo ready for Codex, Claude Code, Cursor, and contributors.

Run Pi coding agent isolated in a Docker Sandbox microVM with a local llama-server as the inference backend

Federated, local-first search for an AI — one query across transcripts, files, knowledge graph, vector store, and the web, fused by trust-weighted RRF. Apache-2.0.

🍌 Autonomous story agent creating illustrated storybooks with Claude Code SDK

A curated collection of Agent Skills — reusable units of intelligence that teach AI General Agents how to perform specific tasks autonomously.

[IEEE T-ITS] A curated list of awesome resources on collaborative (cooperative) perception: papers, datasets, code, and benchmarks for multi-agent and V2X perception in autonomous driving.

Learn Claude Code — 基于源码的完整技术分析文档集，15章深度解析 Agent Loop、工具系统、权限系统等核心机制

Open-source GUI for Claude Code, Codex, and GitHub Copilot CLI. Run coding agents in one workspace with live sessions, code diffs, and a shared browser.

A curated list of the OpenClaw family — the open-source AI agent ecosystem. From the 150K-star original to IoT chips, Rust frameworks, and security tooling.

Shared workspace for AI agents and human developers. See, audit, and review every change AI makes to your codebase. Multi-language graph (Python/TypeScript), MCP-native, local-first.

Unity-aware AI infrastructure for Claude Code — a macOS app that serves a knowledge graph of your Unity project over MCP: 32 tools that let your agent know your project, not just grep its files.

Tandem lets you connect and control any coding agent—including Claude, Codex, or Hermes—across different devices. It allows all your agents to coordinate with each other seamlessly, whether you're working from a browser, a terminal, or any other client.

Herdr and tmux sibling AI sessions: pi, Claude Code, Codex, Copilot, OMP and Antigravity can message and spawn each other while keeping their own auth, tools and transcripts.

My collection of Claude Code Agents & Skills for web and iOS Development

Data2prompt is a powerful cli tool that compiles your entire data-heavy project into a single AI-optimized file. Handles CSVs, Excel, SQLite & notebooks like data, not raw text dumps. data2prompt fits into the Context Engineering phase of the data science lifecycle, bridging the gap between local development and LLM reasoning.

Docker image to run a self-hosted Kokoro TTS server with an OpenAI-compatible audio speech API. 50+ voices across 9 languages, streaming support, all major audio formats, NVIDIA GPU (CUDA) acceleration, offline mode, and persistent model cache. Multi-arch: amd64, arm64.

Open & nano reimplementation of Sakana Fugu. A repo-native, multi-agent coding loop powered by 9+ LLMs (isolated via Claude Code) and an independent Codex reviewer. Lightweight, bounded, and self-improving (Self-Harness)—no coordinator training required.

🐋 Cloud and AI-native multi-agent orchestration platform for Kubernetes

A working repository for designers who use Claude as a collaborator. Skills, principles, project templates, and a career vault.

Semark converts PDFs, Office files, HTML, and images into RAG-ready semantic Markdown using MinerU and VLM/LLM review, with chunks, source maps, and automatic document splitting.

🐟 闲鱼/Goofish 智能监控控制台：AI 大模型选品 + 收藏降价·售罄即时提醒，全本地运行不外传。Xianyu second-hand deal monitor — LLM-filtered picks, price-drop & sold-out alerts, dead-link detection. Playwright · FastAPI · React · Docker.

纯本地 AI 编程桌面工具 · 由 DeepSeek 驱动 · 一个 API Key 即用

Open-source local agent runtime for CTRL NODE — the CLI bridge that pairs your machine with Claude Code, Copilot, Gemini, Cursor and more, and executes tasks from the CTRL NODE dashboard.

Open-source AI operations platform for cross-border commerce: product collection, drafts, AI content optimization, publishing, and order/inventory workflows.

Blind A/B testing for AI agent skills, .cursorrules, CLAUDE.md, and system prompts. Detects placebo and harmful instructions.

gguf merge programming code generation code codeqwen

Evidence-first website copy-remix skill for agent workflows

Agent-SDK without CLI dependencies, as an alternative to claude-agent-sdk, completely open source

Mission control for every Claude Code session on your machine.

Observable diagnostics, failure attribution, metrics, and no-key replay for multi-agent runtimes.

Redeem your Codex banked rate-limit reset from the command line — no UI required.

Autonomous red teaming and evidence-backed security evaluation for live AI agents.

TELL: Test-time Experiential Lifelong Learning – a single LLM agent that learns from experience at test time, achieving 43.9% on ARC-AGI-3 with no training.

Agent skill for AI coding assistants — review and build production-grade Claude agents following SDK best practices

Nano Claude Code - A lightweight AI code agent in Python, inspired by Claude Code and Kimi-CLI. Now you can give it a try and enjoy it plz~~

An agentic AI tool for searching, screening, and ranking NIH funding opportunities based on research interests and proposal relevance.

My LLM Agent Learning

Agent-to-agent messaging for Claude Code sessions on the same machine.

macOS 菜单栏工具，聚合显示多个 AI 编程/Agent工具（Claude Code / Codex / Cursor / Qoder / 悟空/ Workbuddy 等）的 token 用量，全本地直读零配置

Forkable shell substrate for a single code repository — DB-backed identity, memory, roadmap, content; harness-agnostic boot. We build the data layer, we rent the harness.

Vulpine: multi-agent vulnerability-development pipeline for Claude Code and OpenCode

A curated library of 137 production-grade skills for Claude and other AI coding agents.

TWICC - The Web Interface for Claude & Codex - A web UI to browse and interact with your Claude Code and Codex sessions. View projects, sessions, conversation history, costs, and run Claude and Codex agents — all from your browser.

Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record can leak even if the agent is adversarial - enforcement lives in trusted code below the model, validated by a runnable attack gallery.

MetaTrader 5 in a real Windows VM (Docker + QEMU/KVM), driven over REST + MCP — programmatic trading plus server-side TA: OHLC bars enriched with RSI/MACD/Bollinger/ADX/VWAP/Ichimoku/Order Blocks/FVGs (primitives, not signals). Typed MCP tools mirror the whole surface. Multi-broker, multi-account. No Wine bullshit.

Code from anywhere.

Production-ready Agent Skills for Codex, Claude Code, OpenCode and OpenClaw | 小红书图文、学习路径、论文解读与 YouTube 播客笔记

Diverter distills Ultra’s delegation strategy into non-Ultra Codex sessions—bringing in the right native specialist while the main thread keeps moving.

一个从零实现 AI Agent 的开源学习项目——每一步都有源码，每一章都能运行，渐进式理解智能体。Build an AI Agent from Scratch — Learn AI Agents step by step with runnable source code.

Agent Skill that turns plain-English scene descriptions into AI-generated 3D models built live in Blender, via Alpha3D + Blender MCP. Works with Claude Code, Cursor, and OpenAI Codex.

Your AI company, in your device. Local-first AI platform: chat, agent crew, memory, and a model store sized to your hardware - from 4GB laptops to big GPUs. One-click setup, fully private.

Desktop & phone notifications for AI coding agents (Claude Code, Codex, Gemini CLI, Cursor). Native VS Code support. Toast + ntfy push. Zero dependencies.

Claude Code skill that delegates implementation to GPT-5.6 Sol via Codex CLI, Claude plans and reviews the diff, Sol writes the code. Multi-model AI coding where the model that wrote the diff never grades it.

Open, self-hosted runtime for real-time voice agents — a Rust recreation of ChatGPT's Advanced Voice Mode (GPT-Live), model-neutral