23,222 repositories keep their own agent configuration in the open: a CLAUDE.md, a rules file, instructions for Copilot, a skill the project runs on itself. These are not mods to install. They are worth reading for what they reveal — what a maintainer thinks an agent gets wrong.
Building agents7,588Claude Code6,636Data and AI3,676Backend and APIs3,496Languages3,402Planning3,219Git and workflow3,187Code review2,745Testing2,374Memory and context2,364DevOps and cloud2,310Writing and docs2,309Research2,295Frontend2,261Security2,021Productivity1,715
All Claude Code31,998 Codex23,222 OpenCode22,170 GitHub Copilot3,550 Gemini CLI1,918 Cursor1,738 Windsurf137 Kiro80
| Project | Stars | Configures | Files | Always loaded |
|---|---|---|---|---|
| yi427/Cloister Terminal-first, profile-governed Apple container environments for Codex and Claude Code wi | 2 | Codex, OpenCode | 1 | 268 tok |
| hannsxpeter/godplans Plan everything before anything. Single-command AI agent skill that emits a complete, audi | 2 | Codex, OpenCode | 1 | 416 tok |
| product-on-purpose/agent-skills-toolkit product-on-purpose.github.io Toolkit and Standard for building, grading, and scaling cross-agent skill libraries (Claud | 2 | Codex, OpenCode | 1 | 1,732 tok |
| Wandererer/codebeacon pypi.org Source code AST analysis and AI context generation — unified multi-framework knowledge gra | 2 | Claude Code, Codex, Cursor, OpenCode | 3 | 1,670 tok |
| Federico-Anastasi/DYLA-ai Develop Your Local Assistant — a local-first AI workspace that turns project work into val | 2 | Claude Code, Codex, OpenCode | 14 | 5,113 tok |
| cmg8431/claude-dejavu npmjs.com 🔮 Claude Code antipattern detector & auto-patcher — learns from session logs, patches CLA | 2 | Codex, OpenCode | 1 | 589 tok |
| ambystechcom/AmbyKit ambykit.ambystech.io Spec-Driven Development framework that helps AI coding assistants build better software — | 2 | Claude Code, Codex, OpenCode | 2 | 1,473 tok |
| bastndev/skills skills.sh Public repository - Agent Skills (🔥) | 2 | Claude Code, Codex, OpenCode | 2 | 748 tok |
| mpaarating/ai-workflow-kit AI-powered workflow automation skills for Claude Code, Cursor, Codex, and more. Automate t | 2 | Codex, OpenCode | 1 | 493 tok |
| php-workx/fuse A local firewall for AI agent commands | 2 | Claude Code, Codex, OpenCode | 2 | 2,996 tok |
| tk-wxy/apc.md A markdown-based memory system for AI agents — stop re-explaining and re-debugging across | 2 | Claude Code, Codex, GitHub Copilot, Gemini CLI, OpenCode | 4 | 644 tok |
| alexander-akhmetov/programmator Autonomous Claude Code orchestrator that executes multi-task plans without supervision. | 2 | Claude Code, Codex, OpenCode | 2 | 483 tok |
| 1shooperman/claude-env brandonshoop.com Claude Environment Manager | 2 | Claude Code, Codex, GitHub Copilot, OpenCode | 5 | 323 tok |
| gshklovs/react-codex-starter Vite/FastAPI starter code template with CI and AGENTS.MD file for reliable agentic develop | 2 | Codex, OpenCode | 1 | 681 tok |
| gsmlg-app/flutter-app-template Flutter Application Template for AI coder | 2 | Claude Code, Codex, Gemini CLI, OpenCode | 3 | 4,874 tok |
| piecioshka/whose-context-wins Poligon do sprawdzania, który plik konfiguracyjny (CLAUDE.md, AGENTS.md, .cursorrules...) | 2 | Claude Code, Codex, GitHub Copilot, Cursor, OpenCode | 5 | 62 tok |
| thinhledev/mixai npmjs.com An orchestrator for multiple coding agents that utilizes a shared, persistent memory syste | 2 | Claude Code, Codex, OpenCode | 6 | 253 tok |
| bartoszarendt/agenticloop Loop engineering for AI coding agents — shaping the whole run, not one prompt: a Markdown- | 2 | Codex, OpenCode | 1 | 1,980 tok |
| burakdede/lorepack Versioned context build system for AI. Compile documents, spreadsheets and project artifac | 2 | Claude Code, Codex, GitHub Copilot, OpenCode | 3 | 4,145 tok |
| daronthedragon/terse One reflex for coding agents: answer first, cut the padding. A single behavioural skill, h | 2 | Codex, OpenCode | 1 | 298 tok |
| arnarg/crushout A PreToolUse Bash tool hook that auto-approves read-only commands | 2 | Codex, OpenCode | 1 | 1,426 tok |
| alexio777/plan-swarm Agent skill: recursive plan with mechanically gated leaves + dependency-safe subagent wave | 2 | Claude Code, Codex, OpenCode | 1 | 239 tok |
| Waqas-Baloch/superterminal superterminal.dev One rulebook for every AI coding agent. Your rules, context and skills apply to Claude Cod | 2 | Codex, OpenCode | 1 | 123 tok |
| trananhtung/translate-book AI skill to translate books (PDF/DOCX/EPUB) into any language using parallel sub-agents an | 2 | Claude Code, Codex, OpenCode | 3 | 2,001 tok |
| mgerdov-igt/ai-cookbook No-nonsense AI Cookbook for Busy Software Engineers | 2 | Codex, OpenCode | 1 | 469 tok |
| zarubinvibe/mnemazine Local-first memory and knowledge ingestion for coding agents, with auditable archives, ret | 2 | Claude Code, Codex, GitHub Copilot, Gemini CLI, OpenCode | 11 | 1,207 tok |
| xhqing/CC-Bridge Claude Code upstream bridge framework (GLM/DeepSeek/MiMo/Kimi/Qwen) — routes Claude Code t | 2 | Codex, OpenCode | 1 | 8 tok |
| kimtth/ms-fabric-skills-dev-starter microsoftlearning.github.io 🌿 Microsoft Fabric Skills - Scaffolding template for Microsoft Fabric development with AI | 2 | Claude Code, Codex, GitHub Copilot, Cursor, OpenCode | 4 | 1,805 tok |
| RSafargalin/sextant Code intelligence for Swift projects: repository map, structural search, semantics. CLI + | 2 | Claude Code, Codex, OpenCode | 2 | 1,524 tok |
| ihatenodejs/agent-exporter npmjs.com Export usage statistics from various harnesses. | 2 | Claude Code, Codex, OpenCode | 2 | 1,755 tok |
| dootask/skills dootask.com A Claude Code skill set for DooTask plugin development: scaffold plugins / publish release | 2 | Claude Code, Codex, OpenCode | 2 | 572 tok |
| Dtdkvn/codex-oi Universal second-opinion bridge for AI coding agents. Pluggable engines (Codex, Gemini) ca | 2 | Codex, OpenCode | 1 | 1,510 tok |
| JoakimCarlsson/bastion Vibe-coded tower-defense game built by an AI agent pipeline (planner → coder → smoke-teste | 2 | Claude Code, Codex, Cursor, OpenCode | 16 | 2,030 tok |
| jeremylongshore/bobs-big-brain-plugin Bob's Big Brain — the governed team brain, as a local-first Claude Code + Cowork plugin. T | 2 | Claude Code, Codex, OpenCode | 2 | 3,141 tok |
| sshisto/mnemo-mcp 🧠 Manage persistent AI memory with hybrid search and embedded sync for open, free, and un | 2 | Codex, OpenCode | 1 | 1,265 tok |
| RBraga01/builder-ai rbraga01.github.io Your AI assistant will skip the eval. This pack won't let it. Drop-in enforcement skills + | 2 | Claude Code, Codex, OpenCode | 2 | 1,083 tok |
| korbinjoe/TeemAI The GUI for Claude Code, Codex, and your CLI fleet — every session in one place, your atte | 2 | Claude Code, Codex, Cursor, OpenCode | 9 | 1,734 tok |
| joymin5655/Agent Multi-runtime AI agent governance harness — Claude Code · Codex · Gemini under one YAML po | 2 | Codex, OpenCode | 1 | 2,048 tok |
| sergekostenchuk/mouse-trail-masking-reveal Mouse Trail Masking Reveal: portable Agent Skills for cursor-trail reveal effects across C | 2 | Codex, OpenCode | 1 | 135 tok |
| jeremylongshore/intent-eval-lab intentsolutions.io Vendor-neutral research umbrella for measuring AI plugin, agent, and MCP server quality ac | 2 | Claude Code, Codex, OpenCode | 5 | 6,428 tok |
| s977043/PlanGate s977043.github.io Gate-based workflow for AI coding agents | 2 | Claude Code, Codex, OpenCode | 2 | 3,646 tok |
| forsvn-labs/transcript-to-epub Turn a YouTube video or raw timestamped transcript into a clean, readable EPUB book — spea | 2 | Claude Code, Codex, OpenCode | 2 | 334 tok |
| sjarmak/coding-agent-workflows Neutral source of coding standards, agent roles, skills, and workflows, rendered to native | 2 | Codex, OpenCode | 1 | 2,928 tok |
| tarunag10/ai-switchboard ai-switchboard.vercel.app Local-first Mac AI work switchboard for Headroom, RTK, Codex, Claude Code, MarkItDown, Pon | 2 | Claude Code, Codex, OpenCode | 3 | 668 tok |
| YankielDBC2/saas-cybersecurity Provider-aware Agent Skill for evidence-based SaaS security audits and regression-safe har | 2 | Codex, OpenCode | 1 | 335 tok |
| Adit-Jain-srm/skill-forge 18 AI agent skills that catch what your agent misses: hidden code debt, false completions, | 2 | Claude Code, Codex, OpenCode | 2 | 1,461 tok |
| xcaeser/work-skill skills.sh Quality-first orchestration for Codex and other AI coding agents. | 2 | Codex, OpenCode | 1 | 1,006 tok |
| thomas-powers-jr/cadence thomas-powers-jr.github.io Verification layer for AI-assisted development. Agents don't get to self-certify — settle | 2 | Claude Code, Codex, OpenCode | 26 | 8,483 tok |
| lazypaul-apps/md-file-monitor Visual monitor for tracking whether AI coding agents checked project instruction files suc | 2 | Codex, OpenCode | 1 | 69 tok |
| RudAns/OblivionisAgent A wired desktop console for the local Claude Code CLI (different group chats point to diff | 2 | Claude Code, Codex, OpenCode | 2 | 2,127 tok |