45,053 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 agents8,576Claude Code6,909Git and workflow3,638Data and AI3,577Backend and APIs3,561Languages3,499Planning3,322Code review2,907Testing2,704DevOps and cloud2,498Writing and docs2,488Memory and context2,373Research2,320Frontend2,317Security2,182Project management1,989
All Claude Code32,564 Codex24,947 OpenCode23,868 GitHub Copilot3,726 Gemini CLI2,040 Cursor1,770 Windsurf203 Kiro117
| Project | Stars | Configures | Files | Always loaded |
|---|---|---|---|---|
| sardonic-labs/fiducial Turn any AI coding agent into a KiCad hardware design assistant - instruction library + st | 2 | Codex, OpenCode | 1 | 1,554 tok |
| g8rswimmer/cogniflow Visual AI workflow automation platform with a drag-and-drop canvas editor, built-in LLM/RA | 2 | Claude Code | 1 | 2,718 tok |
| beast-bruiser/theartifact-cli theartifact.art CLI to interact with the-artifact-art models and workflow. | 2 | Claude Code | 1 | 2,482 tok |
| howells/scaffold scaffold-drab-eta.vercel.app Opinionated documentation scaffold and project baseline for Howells TypeScript apps, packa | 2 | Claude Code, Codex, OpenCode | 3 | 424 tok |
| webteractive/zetty GUI terminal multiplexer built on libghostty with AI agent detection | 2 | Claude Code, Codex, OpenCode | 3 | 7,884 tok |
| yunze7373/openclaw-termux OpenClaw Termux Fork - Personal customizations | 2 | Claude Code, Codex, GitHub Copilot, OpenCode | 3 | 6,822 tok |
| krayin/agent-skills Krayin CRM Agent Skills | 2 | Codex, OpenCode | 1 | 807 tok |
| Mneno-AI/Mneno A memory layer for AI agents and applications that does more than just “remember”; it pres | 2 | Codex, OpenCode | 1 | 3,120 tok |
| jmfloreszazo/agent-eval-poc Four deployable reference scenarios to govern GitHub Copilot and custom AI agents like any | 2 | Claude Code, Codex, OpenCode | 4 | 884 tok |
| xuanhieu2611/Rivet rivet-dev.vercel.app Autonomous software engineering platform. Submit an engineering job, watch a sandboxed AI | 2 | Claude Code, Codex, OpenCode | 2 | 16,769 tok |
| LuckyOneTwoThree/keel A keel for human + AI project management. Markdown as the database, Git as the audit log, | 2 | Claude Code, Codex, Cursor, OpenCode | 3 | 1,924 tok |
| xmanatee/pub pub.blue | 2 | Claude Code, Codex, OpenCode | 3 | 421 tok |
| tmusser/chart-contract Claim-first chart auditing for Python: validate the data, claim, visual choice, provenance | 2 | Codex, OpenCode | 1 | 2,276 tok |
| jeremy-jm/lando This is a dictionary project that is compatible with all platforms, aims to explore the ap | 2 | Claude Code | 1 | — |
| cueapi/cueapi-cli cueapi.ai Official CLI for CueAPI. Coordinate and verify agent work from the terminal. | 2 | Claude Code | 1 | 622 tok |
| weill-labs/hgm Clean-room reimplementation of the Huxley-Gödel Machine (arXiv:2510.21614): a self-improvi | 2 | Claude Code, Codex, OpenCode | 5 | 1,526 tok |
| ssf0409/tracelens ssf0409.github.io Evaluation and regression-testing framework for AI agents, published in pypi.org/project/t | 2 | Claude Code | 1 | 4,075 tok |
| planailabs/ui.plan.ai ui.plan.ai plan.ai's agent-generated UI streams | 2 | Claude Code, Codex, OpenCode | 9 | 1,678 tok |
| bbonin023/workspace-memory A documentation convention for capturing project state in a way humans and AI agents can b | 2 | Codex, OpenCode | 1 | 1,832 tok |
| aspose-email/agentic-net-examples products.aspose.com Agent-generated C# examples for Aspose.Email for .NET - compiled, executed, and validated | 2 | Codex, OpenCode | 1 | 1,018 tok |
| OrionArchitekton/proctor danmercede.com Proctor: an agent that QAs other agents. UiPath AgentHack 2026, Track 3 (UiPath Test Cloud | 2 | Codex, OpenCode | 1 | 550 tok |
| GovOn-Org/GovOn govon-org.github.io Dong-A University field mirror-type linked project | GovOn Agentic Shell | Collaboration w | 2 | Codex, GitHub Copilot, OpenCode | 2 | 923 tok |
| Michelia-L/AI-WealthPilot A modern wealth management & portfolio quant engine. Features Mean-Variance Optimization ( | 2 | Codex, OpenCode | 1 | 2,500 tok |
| emrecengdev/codex-chief Budget-disciplined multi-agent setup for OpenAI Codex — cheap orchestrator, tiered subagen | 2 | Codex, OpenCode | 1 | 603 tok |
| BombyxLabs/codex-harness-maintenance bombyxlabs.com Report-only harness maintenance for Codex projects: reduce context bloat, token spend, and | 2 | Codex, OpenCode | 1 | 612 tok |
| Cyrus580529/Attest Trustworthy, self-verifying web agent core — proposes actions, the harness validates real | 2 | Claude Code | 1 | 13,350 tok |
| escoffier-labs/agentpantry brigade.tools Your agent machine needs your logins. Pantry ships them sealed. Opt-in AES-256-GCM browser | 2 | Codex, OpenCode | 1 | 1,257 tok |
| mdjakkariya/jacky mdjakkariya.github.io AI Agent for your daily activities and coding task 👨🏻💻 via CLI / chat / voice | 2 | Claude Code | 1 | 2,821 tok |
| aibrohim/react-frontend-template Production-ready React 19 + Vite 7 template with authentication, role-based access control | 2 | Claude Code, Cursor | 2 | 5,001 tok |
| risk-ai/regulator.ai regulator.ai Vienna OS — The execution control layer for autonomous AI systems. Cryptographic warrants, | 2 | Codex, OpenCode | 1 | 1,839 tok |
| dharma-okf-foundation/dharma-okf cloud.google.com The first OKF bundle outside data catalogs — demonstrating the format's power for Dharma k | 2 | Claude Code | 1 | 2,067 tok |
| ametel01/agentreceipt ametel.dev AgentReceipt is a local evidence sidecar for AI-assisted coding sessions. It records sessi | 2 | Codex, OpenCode | 1 | 572 tok |
| ShanHaiFish/dsh-plugin-security-review DSH dynamic Cordis plugin installation security review gate (DeepSeek Harness plugin insta | 2 | Codex, OpenCode | 1 | 1,751 tok |
| vdittgen/arandu Privacy-first AI assistant that runs entirely on your machine. Local Ollama inference, emb | 2 | Claude Code | 1 | 2,103 tok |
| bossandy123/debate-agents AI Talkshow—AI-powered intelligent debate | 2 | Claude Code | 11 | 470 tok |
| agentsfolder/agents-cli agentsfolder.github.io A CLI that projects canonical .agents/ configuration into agent-native surfaces. | 2 | Codex, OpenCode | 1 | 1,484 tok |
| 0SxD/archive-openbrain A brain harness for large language models. Long-term memory, Trinity dialectic, research c | 2 | Claude Code | 1 | 834 tok |
| hatyibei/shingan AI Agent Workflow Static Analyzer — detect infinite loops, cost explosions, PII leak paths | 2 | Claude Code, Codex | 3 | — |
| karandangi123/leadforge-ai leadforge-ai-weld.vercel.app Human-in-the-loop AI RevOps platform for lead research, website audits, outreach drafting, | 2 | Codex, OpenCode | 1 | 74 tok |
| vpeetla-ai/aegisai-enterprise-agent-platform aegisai-enterprise-agent-platform.vercel.app Enterprise agent governance control plane — AI Gateway, HITL, OPA policy, signed audit, Fi | 2 | Claude Code, Codex, Cursor, OpenCode | 14 | 389 tok |
| ykdojo/antigravity-cloud-run Run multiple Antigravity CLI (agy) sessions in isolated Docker containers, locally or on C | 2 | Claude Code, Codex, OpenCode | 2 | 945 tok |
| gvruz80-stack/TermEx-AI Terminal Expanded AI [TermEx AI] is a full screen Terminal Emulator built for making AI CL | 2 | Claude Code | 3 | 1,162 tok |
| Lexideck-Technologies/Lexideck2026 A six-agent creative and research federation for OpenAI Codex, with 19 reusable skills and | 2 | Codex, OpenCode | 20 | 876 tok |
| Fox-River-AI/agent-atlas-studio Erwin-style visual modeler for agentic AI platforms — orchestrator, tasks, agents, MCP too | 2 | Claude Code | 1 | 1,889 tok |
| Arclude/Arterm-CLI arterm.dev Terminal AI coding agent that runs local models. Connect to Ollama, any OpenAI-compatible | 2 | Claude Code, Codex, OpenCode | 2 | 641 tok |
| QWED-AI/qwed-infra docs.qwedai.tech Deterministic verification layer for infrastructure as code. Verifies AWS IAM policies, ne | 2 | GitHub Copilot | 1 | 205 tok |
| neilshermes-project/hermes-memory-hybrid One-command hybrid memory upgrade for Hermes Agent (FTS5 + sqlite-vec + embeddings). Fully | 2 | Codex, OpenCode | 1 | 307 tok |
| Endokelp/Should-I-build Multi-agent market validation skill for Claude Code and Cursor — BUILD / CONDITIONAL / PIV | 2 | Codex, OpenCode | 1 | 464 tok |
| mochow13/keen-agent Terminal agent harness built from a coding-agent foundation — plug in your own agent via Y | 2 | Claude Code, Codex, OpenCode | 3 | 321 tok |
| w48669664-bit/colla-agent Local-first, observable Codex × Antigravity collaboration with bounded reviews and replaya | 2 | Codex, OpenCode | 1 | 425 tok |