Code agents

Agent skills for long-running execution, high-recall code review, and measurable self-improvement loops.

A vision backend for AI Agents — self-healing Claude Code Agent Skill. Routes images to Mimo/OpenAI/Ollama vision API.

Official codebase for the paper "WorldMemArena: Evaluating Multimodal Agent Memory Through Action–World Interaction"

Unity-aware AI infrastructure for Claude Code — a knowledge graph + 88 MCP tools that let your AI 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.

Garden-citizen dispatch substrate + meta-bridge: make Claude Code, Codex, Antigravity, and pi sessions garden-addressable peers — with a Claude-first ACP plugin (Cortex/vendor ACP backends are future lanes).

53 retro HTML/CSS UI themes for AI-assisted frontend development. Windows 95, CRT, vaporwave, DOS, arcade, mainframe, and more.

Autonomous AI call agent for GoHighLevel: calls leads, qualifies them, books appointments, and updates your CRM on autopilot. ~300 calls a day, unattended. Built in n8n with Vapi, Twilio & OpenAI.

Let's Qitos! A torch-like agent-native framework for researchers.

Spec Kit writes WHAT. Superpowers enforces HOW. Thin handoff bridge for Codex and Claude Code.

布谷 · 给长跑 coding agent 的「声音信标」:防止 Mac 休眠,每次状态变化用声音提醒,一键跳回正确的终端标签页。 | Bugu — a sound beacon for long-running coding agents: keeps your Mac awake, chirps on every state change, and jumps you back to the exact terminal tab.

Cognitive memory for LLM agents — local-first, and shareable across a team through the cloud.

Ultimate collection of Codex tips, tricks, and workflows that you can use to master Codex CLI in minutes.

Obsidian 图文导出插件：将 Markdown 笔记一键生成小红书 / RedNote、抖音图文、Bilibili 图文等社媒卡片，支持主题、封面和批量导出。

Открытый бенчмарк LLM: какая нейросеть лучше пишет код 1С:Предприятие (BSL). Объективная оценка LLM по методике SMOP с реальным исполнением в 1С — Claude, GPT, Gemini, DeepSeek, YandexGPT, GigaChat.

Catch silent failures in AI agents before your users do

⚡ The everything Zotero MCP server: complete Zotero Web API v3 + desktop local API for Claude & any MCP client. Search, safe writes, add-by-DOI, CSL citations, hybrid semantic search, and a scholarly-context graph. TypeScript, local-first.

🏛️ Mnemosyne OS — A cognitive memory OS for AI agents. Magic Memory Palace (taxonomy × archive-no × tome cards × 3-channel summon), TMT distillation, knowledge graph, WIKI KB, Hermes-native. Self-hosted. / 认知型记忆操作系统 · AI 长期记忆宫殿 · 自托管

Free MCP servers + Claude Code Skills for MSP tools - PSA, RMM, backup, M365. Local-first: your data never leaves your network.

The curated list of tools, standards, and resources for governing AI agents in production: policy enforcement, audit trails, EU AI Act compliance, MCP and Claude Code security.

Use your OAuth subscriptions or API keys in any coding agent, Factory Droid, Cursor, and more. Tunnel Agent handles provider authentication and routes requests through CLIProxyAPI or 9Router.

Use Sleepwalker to access AI visibility data (prompts, answers, citations) and Content Intelligence insights (trends, opportunities). Available via MCP, API, CLI and Sleepwalker App.

Semantic F# for AI coding agents (Claude Code · Cursor · Copilot · Codex) — an MCP server that gives LLMs real FCS/FSAC compiler intelligence (find, check, types, rename) instead of grepping F#.

🔥 CogitoAgent — An autonomous AI agent, cloud-driven but locally executed. It continuously thinks, automatically analyzes, and integrates practical tools like file management, code execution, web search, and database operations — secure and efficient.

A/B test agent skills with blind grading + bootstrap CIs - does your SKILL.md actually improve task performance?

transformers safetensors llama text-generation code granite

transformers safetensors qwen3_moe text-generation mlx conversational

gguf text-generation base_model:Qwen/Qwen3-Coder-Next base_model:quantized:Qwen/Qwen3-Coder-Next license:apache-2.0 endpoints_compatible

gguf glm codegeex thudm text-generation zh

spec-builder + harness-bootstrap. Specs both you and the AI understand. An agent team tailored to your repo, derived from your contract and your code, not a kit that installs everything. Guardrails that block, not advise.

Open-source GraphRAG domain knowledge layer: LLMs continuously distill scattered skill docs into a queryable, evolvable, human-editable two-layer knowledge graph — versioned, hot-swapped and rolled back via a GitOps pipeline like software releases, and served to AI agents over MCP. RAG, but with relational structure.

Persistent memory for AI coding agents — SQLite + FTS5 + vector search in a single file

🔁 Write loops, not prompts. Free skill that generates overnight autonomous AI runs for Claude Code, Codex & Cursor: master prompt, plan, verifier, one command. By AI Guy Official.

Archived first-generation Claude Code framework with agents, commands, rules, templates, and MCP integrations.

175+ pip-installable tool integrations for the ModuleX runtime — one pydantic manifest contract, entry-point discovery, async httpx tools

Schema-locked, drift-checked SEO automation as a Claude Code plugin — 49 skills, 29 slash commands, auditable agent workflows (GSC, DataForSEO, Screaming Frog, Scrapling)

Cyberful is an open-source application-security workbench for discovering, exploiting, verifying, and remediating vulnerabilities.

Multi-agent coding template — persistent memory, spec-driven workflow, skills and security hooks for Claude Code, Cursor, Windsurf, GitHub Copilot and any AGENTS.md-reading tool. No dependencies.

AI-Powered Multi-Agent Crypto Investment Analysis Platform

An agent-native workspace for quantitative research: an in-terminal notebook, a high-performance time-series database, and a high-throughput backtesting engine. Written in Rust.

Run one AWS CLI command across hundreds of AWS accounts in parallel. MCP built in for AI agents.

An AI coding agent that works where your code lives, including remote servers over SSH.

The identity provider for AI agents — registry, scoped delegation, in-path MCP enforcement, instant revocation, tamper-evident audit.

Source code for our paper: "Think Fast and Slow: Step-Level Cognitive Depth Adaptation for LLM Agents".

ClawGuard is a comprehensive security toolkit designed to mitigate risks associated with autonomous agents, such as OpenClaw and other LLM Agents.

My near-autonomous workflow for coding agents

A modular simulation framework for controlling multiple drones in a PX4-based environment. It features ROS 2 integration, G-code trajectory planning, multi-agent communication, and a Dockerized setup for easy deployment and visualization.

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

Simple code sandbox supporting jupyter notebook style code execution. Used for agent training

Desktop command center for managing multiple AI coding agent sessions across codebases. Multi-agent orchestration via MCP.

Code for ICML25 paper "HYGMA: Hypergraph Coordination Networks with Dynamic Grouping for Multi-Agent Reinforcement Learning"

基于 EdgeOne 的自托管免费微信推送与标准 MCP Server，支持 Agent / IDE、Webhook 风格 API、多应用和订阅群发。Self-hosted free WeChat push service and standard MCP server on EdgeOne, with Agent/IDE support, webhook-style API, multi-app support, and subscription broadcast.

Rapidly build a multi-Agent management dashboard based on Claude Code, integrating daily work and remote IM collaboration. It includes Claude Code monitoring, channel configuration and message notifications, with a Codex GUI optimized for non-technical users.

An MCP server that lets AI agents read and draw on the Figma canvas — safely. Paired with a Figma plugin, it provides structured verification, safe-by-default drawing, and resilient multi-window support for Claude Code, Codex, Cursor, and other MCP clients.

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

VS Code Extension for searching, installing and managing Agent Skills (SKILL.md) for GitHub Copilot, Claude Code and AI coding assistants

🔧 LRA - AI Agent长任务超稳定执行框架 ⚡ 轻量 | pip一键安装｜突破上下文限制 📌 7种状态管理 | Long-Running AI Agent Framework • One-click pip install｜Break context barriers • 7-status management. 5.0全面支持CLAUDE/OPEN CLAW/OPEN CODE/CUSOR/TREA 等全部Agent编程工具，仅需运行 lra init命令，就会生成Agent.md/CLAUDE.md/lra.md. 你的agent会自动学会如何使用

Korean translation of "Harness Engineering: From Claude Code Internals to AI Coding Best Practices" — Claude Code 소스 코드를 체계적으로 분석하여 AI Agent 시스템의 핵심 엔지니어링 패턴을 추출한 기술서의 한국어

Research Harness for Claude Code. Keep your agent grounded in context, not lost in vibe coding.

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.

Long-task-oriented video analysis skill for Claude Code (/vea) — resumable staged execution, automatic parallel scoring, OpenClaw orchestration, audiovisual analysis with agent handoff

Persistent memory for AI coding agents. Your agent learns your codebase the way a senior engineer would — what files go together, what you usually edit next. Works with Claude Code, Cursor, Cline, Continue. 100% local. (Side effect: 12-50× cheaper code questions.)

Free Gemini web interface proxy for OpenCode - Use your free Gemini subscription as an OpenAI-compatible API with full tool calling support

Fast OpenAI SDK 是一个专为 Java 开发者设计的轻量级、高性能 SDK，旨在提供与 OpenAI 及 Jina API 完美兼容的客户端调用接口。它不仅具备极简的接入体验，还内置了流式响应（SSE）、多 API Key 负载均衡、请求重试、视觉能力以及大模型函数调用（Function Calling）等核心企业级特性。

Agentic Framework for Synthesizing CodeQL Queries

Converts skill/framework documentation into compressed AGENTS.md indexes for AI coding agents. Based on Vercel's research showing AGENTS.md achieves 100% pass rate vs 53% baseline.

The declarative reasoning layer for AI agents. A portable, framework-agnostic file format for specifying how an agent thinks — strategy, step DAGs, contracts, quality gates, and fallback policies — declared in YAML rather than hardcoded in Python.

Community-driven catalog of ready-to-use agents for the Security Copilot portal.

An autonomous, 100% local Agentic GraphRAG system for codebase intelligence. Paste any GitHub URL or local directory to chat with your code. Powered by LangGraph, NetworkX, and Ollama (qwen2.5-coder).

Drop-in skills that stop AI coding agents from shipping garbage — kill-slop, security-review, PR hygiene

🧙 Grimoire is a skills package manager for AI agents and best practices — 1000+ skills across 26 domains (engineering, law, finance, design, health, and more), each sourced from a real institution or study. Install once, and your AI routes to the right skill automatically.

Resources for building and deploying agents using Claw and agent harness in Microsoft Foundry — patterns for agent orchestration, testing, and production deployment. From Microsoft Build 2026.

Mission control for parallel Claude Code & Codex sessions — a live grid of coding agents that glow when they need you, across multiple accounts, resumed after restart.

Open, MIT-licensed food & nutrition science research skills for Claude Code, Codex, and MiniMax Agent — multi-agent literature/systematic review, journal-aware writing, peer review, and figures, plus author-guideline skills for 150+ journals. Initiated by the Food Science Group, University of Melbourne.

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

Collection of Hermes Agent profiles — persona-driven AI agent configurations for orchestration, coding, and research.

Claudex is an open-source AI coding agent with multi-provider support (OpenAI, NIM, Ollama, and more), featuring a smart routing system, Telegram integration, and full Claude Code tooling — all powered by any model you choose.

MCP server and CLI for connecting AI agents, OpenClaw/Hermes workflows, local automation, and coding harnesses to your RemNote knowledge base via the RemNote Automation Bridge.

Clint is the AI Era Outreach System. It is a Python library and CLI (Command Line Interface) designed for production-grade lead discovery, evidence-based personalization, and safe outreach automation. It bridges the gap between raw data scraping and personalized human-like communication.

Production-ready multi-agent platform delivering orchestration, session governance and audit capabilities. Stack: Go/Python/Rust microservices, multi-type data storage, Next.js admin UI, automated CI/CD pipelines, AI code review triggered via PR comment mentions.

Run Claude Code with GPT models through Raine's claude-code-proxy.

The AI Autonomous SOC & Purple-Team Engine

Local-first AI agents for your Obsidian vault.

Six security outcomes an enterprise must achieve to run AI agents, and the controls that evidence them. A draft for public comment.

Open-source infinite vector canvas and AI Design Agent — shape, text, media, and path editing with realtime collaboration; self-host with Docker Compose.

LongCat-Flash-Chat safetensors text-generation transformers conversational custom_code

Typography skill for AI agents - audit and produce publication-grade text. Butterick's Practical Typography for your coding agent.

Fix agent reliability issues in CI across Claude Agent SDK, OpenAI Agents SDK, Google ADK, MCP, LangChain, LangGraph, CrewAI, AutoGen AG2, Pydantic AI, and Vercel AI.

A local-first permission firewall and approval layer for AI agent tool calls.

A Perplexity API MCP server that unlocks Perplexity's search-augmented AI capabilities for LLM agents. Features robust error handling, secure input validation, and transparent reasoning with the showThinking parameter.

An open-source Coding Harness Platform for long-horizon, multi-agent software delivery.

Self-hosted, OpenAI-compatible speech server: faster-whisper STT and Piper/Kokoro TTS. A local voice backend for Open WebUI, CLIs, and custom agents.

Collection of useful skills for Claude Code - AI agents, frontend design, and more

A personal AI operating system built on OpenClaw/Claude Code, MCP servers, and structured markdown. Feedback RL, voice cloning, Screenpipe integration.

Claude Code skill for planning, executing, and measuring digital marketing campaigns. Includes UTM tracking, content strategy, email sequences, social media tactics, brand voice guidelines, and analytics frameworks.

A comprehensive Claude Code plugin for Frappe Framework development, providing tools, commands, agents, and Skills to streamline your Frappe application development workflow.

Claude Code, Codex, project skills, plugins, collections, and agents in the menu bar on macOS.

Lightweight agent framework for research PowerPoint generation, editable PPTX delivery, and rendered QA

AI agent workflow platform — visual flow orchestration, drag-and-drop forms, knowledge base RAG, multi-model LLM, digital workers, Tauri desktop & DingTalk bot. Open source, self-hosted.