43,652 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 |
|---|---|---|---|---|
| calven-ai/demoverse Grow a fake company's entire sales history across CRM, calls, email and Slack that moves e | 4 | Claude Code, Codex, OpenCode | 7 | 3,794 tok |
| Abhillashjadhav/AI-PM-essential-skills Installable Claude Code plugins for AI product evaluation, model routing, guarded loops, a | 4 | Claude Code, Codex, OpenCode | 3 | 1,227 tok |
| saitarrun/Sdlc-ai-workflow npmjs.com Claude Code plugin for end-to-end SDLC automation with role-specific agents, reusable skil | 4 | Claude Code | 1 | 1,213 tok |
| netresearch/netresearch-branding-skill netresearch.github.io Agent Skill for implementing Netresearch brand guidelines - colors, typography, components | 4 | Codex, OpenCode | 1 | 808 tok |
| truman-t3/industrial-design-portfolio-skill truman-t3.github.io Evidence-first industrial design portfolio Skill for Codex, Claude Code, Cursor, Gemini CL | 4 | Claude Code, Codex, Gemini CLI, OpenCode | 3 | 7,821 tok |
| crimx/skills A collection of agent skills I use in real work. | 4 | Codex, OpenCode | 1 | 490 tok |
| ca-who-codes/ultimate.UIUX.design.skills One-stop design context pack for AI coding agents — two tracks: interfaces (web/apps, comp | 4 | Claude Code, Codex, OpenCode | 2 | 1,707 tok |
| kitepon/Throughline npmjs.com Cut ~90% of Claude Code's context usage while keeping nearly all the memory | 4 | Claude Code, Codex, OpenCode | 4 | 21,134 tok |
| amlei/amlei-resume An Agent Skills package for writing resumes, running in AI assistants such as Claude Code, | 4 | Codex, OpenCode | 1 | 995 tok |
| rhein1/fable5-codex OpenAI Codex plugin for evidence-first code review, audits, fact checks, run contracts, an | 4 | Codex, OpenCode | 1 | 372 tok |
| Beat-YT/omnivox-mcp Self-hosted bridge to Omnivox (Quebec CEGEP portal): MCP server, REST tool gateway with Op | 4 | Claude Code | 1 | 2,489 tok |
| PralhadYadawad/avvarre avvarre.com AI context that survives the chat window. Persistent repo memory, impact analysis, and ins | 4 | Claude Code, GitHub Copilot, Cursor | 17 | 2,387 tok |
| aderaaij/loopback-training-server Self-hosted backend for Loopback – Run Coach: workout tracking, Apple Watch training queue | 4 | Claude Code | 1 | 8,389 tok |
| monkfromearth/intermind npmjs.com An MCP server that lets Claude Code, Codex, Cursor, Cline, Windsurf, and any other MCP-spe | 4 | Claude Code | 1 | 2,820 tok |
| MontoyaAndres/ganju ganju.ai Ganju.ai is a no-code tool for creating fast and scalable MCP servers. | 4 | Claude Code | 1 | — |
| deploymenttheory/windows-mcp-server A MCP server that bridges AI agents to the Windows desktop. UI Automation, synthetic mouse | 4 | Claude Code | 1 | 8,704 tok |
| agentdecksdk/agentdeck agentdecksdk.com AgentDeck SDK: the execution layer for agentic software. Agents, tools and workflows as pl | 4 | Claude Code | 17 | 2,000 tok |
| ASK191225/bugbounty-kit Production bug bounty operations kit — LLM brain, token scanner, HackerOne MCP, RAG pipeli | 4 | Claude Code | 2 | 5,395 tok |
| Xerrion/servicenow-platform-mcp A developer & debug-focused MCP server for ServiceNow - platform introspection, change int | 4 | Codex, OpenCode | 1 | 30 tok |
| hexproofdev/oring Unprivileged, self-applied Landlock + seccomp confinement wrapper — no root, no daemon, on | 4 | Claude Code | 1 | 1,181 tok |
| everclear077/dsh-progressive-tools Progressive tool discovery for DeepSeek Harness — tiny stable surface, searchable catalog, | 4 | Claude Code, Codex, OpenCode | 2 | 944 tok |
| lizard-build/lizard-cli lizard.build Deploy and manage apps on Lizard from your terminal — upload and deploy code, stream logs, | 4 | Codex, OpenCode | 1 | 1,516 tok |
| agno-agi/agentos-render docs.agno.com AgentOS: Serve agents over API, MCP, and chat interfaces like Slack | 4 | Claude Code, Codex, OpenCode | 11 | 15,298 tok |
| MuhammadUzair1/Vibe-Engineering-SMIT vibe-engineering-smit.vercel.app | 4 | Claude Code | 1 | — |
| xsourabhsharma/ai-security-audit-pro Universal security-audit plugin and CLI engine for AI agents with OWASP-mapped reports for | 4 | Claude Code, Codex, Gemini CLI, OpenCode | 3 | 753 tok |
| guicybercode/Jig Terminals on a canvas. A local-first desktop workspace for coding-agent CLIs, parallel PTY | 4 | Codex, OpenCode | 1 | 1,134 tok |
| miguepoloc/weather_station | 4 | Claude Code | 1 | — |
| cosmo-autom8s/ai-executive-assistant Your personal executive assistant that lives in your terminal. Built on Claude Code. Manag | 4 | Claude Code, Codex | 25 | — |
| MilindGaharwar/fettle pypi.org Open-source trust layer for AI coding agents: in-session quality, delegation-safe policy, | 4 | Claude Code, Codex, GitHub Copilot, OpenCode | 18 | 250 tok |
| sernst/skills My AI Agent Skills. A nix of ones I've developed myself and others I've adapted from other | 4 | Codex, OpenCode | 1 | 446 tok |
| aakashg/pm-github-workflow-repo Practice repo for PMs: learn Git/GitHub for your AI artifacts (skills, CLAUDE.md, evals, a | 4 | Claude Code | 1 | 380 tok |
| staticaland/skills Ceremonial grade agent skills 🫖 | 4 | Claude Code, Codex, OpenCode | 7 | 949 tok |
| aemcoder/skills Set of skills for AEM Coder Stack | 4 | Claude Code | 1 | 1,546 tok |
| Netcracker/qubership-ai-packages | 4 | Claude Code, Codex | 9 | — |
| howells/skills | 4 | Claude Code, Codex, OpenCode | 2 | 2,055 tok |
| franciscorizzo/GxObjGen-install | 4 | Claude Code | 2 | 1,361 tok |
| RichardHightower/reliable-agentic-lab Lab repo for Engineering Reliable Agentic AI Systems (Packt workshop, 29 August 2026). Tic | 4 | Claude Code, Codex, OpenCode | 6 | 795 tok |
| zell12/claude-certified-architect-study-labs | 4 | Claude Code | 1 | — |
| snyk-labs/snyk-vulnbench | 4 | Claude Code, Codex, OpenCode | 12 | 2,521 tok |
| walterwang0x01/tech-learning-and-projects walterwang0x01.github.io A collection of personal technical study notes and practical projects | 4 | Kiro | 20 | 15,808 tok |
| Aidas-dev/k8s-agent-skills git.kubexa.tech Agent skills for Kubernetes cluster operations — Cilium, Talos, Flux, Rook-Ceph, CNPG, Git | 4 | Codex, OpenCode | 1 | 333 tok |
| yandong2023/java-backend-team An AI Skill configuration package for enterprise Java backend teams, designed to help deve | 4 | Codex, OpenCode | 2 | 1,789 tok |
| accordproject/concerto-rust Implementation of Concerto runtime in Rust | 4 | Codex, OpenCode | 1 | 279 tok |
| Thomaszhou22/self-discover-skill SELF-DISCOVER: AI agents self-compose reasoning structures. Based on Zhou et al. (2024), N | 4 | Claude Code, Gemini CLI | 2 | 99 tok |
| vanities/social-skills Skill-driven social media automation across IG / LinkedIn / X / Pinterest / TikTok via Cla | 4 | Claude Code, Codex, OpenCode | 2 | 6,807 tok |
| NetApp/gcnv-mcp-server MCP server for provisioning and managing NAS and iSCSI volumes using Google Cloud NetApp V | 4 | Claude Code, Codex, Gemini CLI, OpenCode | 3 | 23,364 tok |
| AEALab/CraftWiki | 4 | Codex, OpenCode | 1 | 178 tok |
| TserenTserenov/ZP Zeroth Principles: transdisciplinary meta-constraints + hierarchy map (0→1→2→3) | 4 | Claude Code | 1 | 440 tok |
| JayMishra-source/Fantasy-Football-AI-CoManager Complete LLM-agnostic fantasy football automation system with ESPN API integration, MCP se | 4 | Claude Code | 1 | 2,641 tok |
| OpenAdaptAI/openadapt-evals pypi.org Windows Agent Arena and GUI-agent evaluation for OpenAdapt. Live support surface. Install | 4 | Claude Code | 1 | 10,026 tok |