How real projects configure their agents

22,170 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
slopstopper/plumb-line slopstopper.github.io Stop uncertain data becoming confident-looking results. plumb-line carries provenance, con 4 Codex, OpenCode 1 1,774 tok
jeltehomminga/figma-design-skills Two composable Claude Code skills for design-to-code fidelity: extract exact specs from Fi 4 Codex, OpenCode 1 786 tok
echennells/sparkbtcbot x.com Agent skill for Spark Bitcoin L2 wallet capabilities including L402 for AI agents 4 Claude Code, Codex, OpenCode 2 3,868 tok
calt13/claude-skills-integration-templates Drop-in Markdown snippets that enable any Agent CLI to discover and use Claude Skills — ev 4 Codex, OpenCode 1 1,789 tok
vcode-sh/dokploy-mcp npmjs.com MCP server for Dokploy. Two tools up front, 524 generated procedures underneath. Real depl 4 Claude Code, Codex, OpenCode 2 769 tok
gryszzz/Kaspa-Ai-Agent-Skill gryszzz.github.io Kaspa sovereign architect engine ; engineering skill for AI agents, with source-gated prot 4 Codex, OpenCode 1 664 tok
Neetx/ai-security-research-radar AI Security Research Radar — AI for security and security for AI 4 Claude Code, Codex, OpenCode 14 2,892 tok
piotrkowalczuk/zordon zordon.io Give every parallel coding agent its own local stack — databases, brokers and services as 4 Claude Code, Codex, OpenCode 5 1,864 tok
subodhkc/ai-appsec Evidence-backed AppSec for AI applications and agents. Powered by HAIEC. 4 Codex, OpenCode 1 2,730 tok
jamestkelly/diffler Generate comprehension quizzes from Git branch diffs for Claude Code, OpenCode, and pull-r 4 Claude Code, Codex, OpenCode 2 743 tok
habinsong/k-fable-opus-claude-md After being shocked by Fable 5's performance and price, someone built a stripped-down md f 4 Claude Code, Codex, Gemini CLI, OpenCode 3 10,000 tok
Zephyrex21/claude-leetcode-helper A Claude Code Skill that transforms LeetCode problems into optimized interview-ready solut 4 Codex, OpenCode 1 3,919 tok
bx33661/oh-my-vul npmjs.com Evidence-first vulnerability research workspace and Skills for Claude Code and Codex. 4 Codex, OpenCode 1 449 tok
5kahoisaac/skillless You don’t need to be skillful — your agent’s got all the skills. Being skillless is totall 4 Claude Code, Codex, OpenCode 2 602 tok
coroboros/agent-skills coroboros.com AI agent skills for Claude Code and compatible agents — workflow, coding, design, Claude C 4 Claude Code, Codex, OpenCode 2 718 tok
MercuriusDream/agent-estate 4 Codex, OpenCode 1 1,251 tok
tuliosousapro/SaaS-blueprint tuliosousapro.github.io BRAINIAC - SaaS Blueprint, NOT an APP. It's a comprehensive playbook and operating system 4 Claude Code, Codex, OpenCode 1 1,777 tok
8Dionysus/aoa-skills Bounded agent-facing skill bundles for coding agents and humans: reviewable execution work 4 Codex, OpenCode 1 755 tok
rolandbrecht/agent-skills A collection of shareable, high-quality AI coding agent skills and workflows. These skills 4 Claude Code, Codex, OpenCode 2 1,084 tok
souljorje/agent-skills Collection of hand-crafted skills 4 Codex, OpenCode 1 322 tok
computerlovetech/skills Agent skills published by computerlove.tech <3 4 Claude Code, Codex, OpenCode 2 229 tok
eduardozf/wire-lang wirelang.net 4 Claude Code, Codex, OpenCode 2 553 tok
pax-k/build-right Agent-agnostic skills that turn a fuzzy app idea into evidence-backed AI execution. 4 Codex, OpenCode 1 770 tok
entr0pia/scoop-skills Scoop Skills: A specialized skill for the AI Agent agent to manage Windows packages using 4 Codex, Gemini CLI, OpenCode 2 1,788 tok
musabekisakov-imj/flutter-design-engineer Agent skills for product-aware, adaptive, accessible, and visually verified Flutter UI des 4 Codex, OpenCode 1 283 tok
gustavo-meilus/superpipelines Loop Engineering for AI coding agents, with real review boundaries. Your AI reviewer canno 4 Claude Code, Codex, Gemini CLI, OpenCode 8 2,954 tok
tmusser/ai-engineering-skills Portable AI-engineering skills for Claude Code, Codex, and coding agents: bounded scope, m 4 Codex, OpenCode 1 385 tok
n-shadloo/git-authoring Agent skill that turns your git diff into clean Conventional Commits — plus on-request fil 4 Codex, Gemini CLI, OpenCode 2 12,346 tok
mpbt-hq/starfleetctl multi agent coordination / command & control toolkit following the Startrek fleet metaphor 4 Codex, OpenCode 1 82 tok
cosgroveb/debian-packaging-agent-skill Debian packaging skill with comprehensive policy, debhelper, and language-specific knowled 4 Codex, OpenCode 1 206 tok
kaltura/kaltura-agents-api-guides kaltura.md Everything AI agents need to master the full Kaltura platform — The Agentic Digital Experi 4 Claude Code, Codex, OpenCode 5 8,819 tok
ishands/upstack An open-source framework that configures AI as a genuine tutor — not an answer machine. 4 Claude Code, Codex, OpenCode 2 2,396 tok
MrSGSA/math-modeling-skill-dify A mathematical modeling Agent Skill and Dify Jiuku multimodal MCP bridge that can be dragg 4 Claude Code, Codex, OpenCode 2 881 tok
matt-bat/agent-command-center Governance-first agent command center for workflow automation, validation, policy, documen 4 Codex, OpenCode 1 636 tok
kelegele/oh-my-pm Oh-My-PM - Product Manager AI Agent Workflow System via Claude Code Skills 4 Claude Code, Codex, OpenCode 2 4,470 tok
vraj-ai/skills Agent skills for running software work as a factory, not a conversation: a four-stage pipe 4 Codex, OpenCode 1 176 tok
Grails125/codex-sol-terra-luna-auto-orchestrator Automatic risk-based task routing and multi-agent orchestration for Codex using Sol, Terra 4 Codex, OpenCode 1 1,598 tok
conallob/o11y-analysis-tools Various static analysis and testing tools for managing PromQL compatible monitoring stacks 4 Claude Code, Codex, OpenCode 2 6,146 tok
mpagot/os-autoinst-distri-opensuse-skills mpagot.github.io OSADO educated clankers 4 Codex, Gemini CLI, OpenCode 2 2,198 tok
VersoXBT/claude-initial-setup 75 skills, 14 agents, 15 commands for Claude Code. Plug-and-play starter kit covering Type 4 Codex, GitHub Copilot, Cursor, OpenCode 3 4,908 tok
augchan42/inkstone sixlines.online 硯台 — Claude Code skills for creative and design workflows. Ink painting prompts from class 4 Claude Code, Codex, OpenCode 2 3,062 tok
deanpeters/product-manager-antagonist-skills Structured dissent skills for product practitioners preparing for consequential decisions 4 Codex, OpenCode 1 7,613 tok
buildinternet/releases-cli releases.sh Agent-friendly CLI for searching the Release Index catalog of release notes, changelogs, a 4 Claude Code, Codex, OpenCode 4 5,147 tok
DaleStudy/skills skills.sh 🛠️ Agent skill. 4 Claude Code, Codex, OpenCode 2 1,621 tok
mdsohaib/screenshot-time-machine Screenshot Time Machine is like a Wayback Machine for your own website, a visual archive o 4 Claude Code, Codex, OpenCode 2 984 tok
hraness/wrench wrench.rip Open-source CLI and TypeScript SDK for precise web capabilities for AI agents: page captur 4 Claude Code, Codex, OpenCode 13 5,061 tok
ni032mas/agent-skills-ru Russian AI agent skills for legal, compliance, business, product and engineering workflows 4 Codex, OpenCode 1 189 tok
joris887/exosuit exosuit.sh Suit up. Build anything. A drop-in engineering organization for Claude Code — deep idea in 4 Claude Code, Codex, OpenCode 13 1,654 tok
natnew/Awesome-Agentic-Engineering natnew.github.io A curated map of agentic AI systems — covering architectures, frameworks, memory, evaluati 4 Claude Code, Codex, GitHub Copilot, OpenCode 7 8,153 tok
arsenelupin14/THEMIS Research workrooms for AI agents: evidence retrieval, citation audit, adversarial review, 4 Codex, OpenCode 1 458 tok