AGENTS.md examples from real projects

strapiAGENTS.md

strapi

strapi/strapi/AGENTS.md

production fixes docs/ # Contributor docs site (contributor.strapi.io); package docs mirror packages/ under docs/docs/packages/ tests/ # Integration, E2E, and CLI test infrastructure ``` The following are the most important packages (not exhaustive

No other project uses it yet

strapiAGENTS.md

strapi

strapi/strapi/docs/AGENTS.md

page. Use the [`contributor-docs` skill](../.ai/skills/contributor-docs/SKILL.md) for the procedure. ## Layout ``` docs/docs/contributing/ # how-to: testing, conventions, writing docs docs/docs/architecture/ # explanations that cross packages docs/docs/packages/ # one folder per package, mirrors packages

No other project uses it yet

usestrixPipes a download into a shellAGENTS.md

strix

usestrix/strix/AGENTS.md

agent skills for step-by-step workflows: ```bash npx skills add usestrix/strix ``` - `penetration-testing-with-strix` — run a headless pentest against code, URLs, domains, or IPs and read results (covers

nektosReads credentialsAGENTS.md

act

nektos/act/AGENTS.md

line in `go.mod` (currently 1.26.0). ## Commands - `make build` — build the binary to `dist/local/act` - `make test` — `go test ./...`, then run the act CLI against a sample workflow (needs a Docker daemon

hakimelAGENTS.md

reveal.js / react

hakimel/reveal.js/react/AGENTS.md

react/src/index.ts`. - Shared types live in `react/src/types.ts`. - The Reveal instance context lives in `react/src/reveal-context.ts`. - Component tests are colocated with their components as `*.test.tsx`. - Test setup remains in `react/src/__tests__/setup.ts`. - The demo

No other project uses it yet

mobyAGENTS.md

moby

moby/moby/AGENTS.md

opening pull requests. It is the source of truth for contribution scope, coding style, testing, commits, DCO sign-off, review, AI usage, and release notes. # Pull requests When opening

No other project uses it yet

ansibleAGENTS.md

ansible

ansible/ansible/AGENTS.md

project. ## General Principles When reviewing code, don't flag issues that `ansible-test sanity` already catches. Focus review effort on things automated checks can't verify. ## Attribution Agents should disclose

No other project uses it yet

ClineReads credentialsAGENTS.md

cline

cline/cline/AGENTS.md

ANTHROPIC_API_KEY`, `CLINE_API_KEY`, `OPENROUTER_API_KEY`); see `apps/cli/README.md`. ### Build / Lint / test - SDK packages (`@cline/shared|llms|agents|core|sdk`) resolve each other through compiled `dist/` (their `exports` point

No other project uses it yet

ClineAGENTS.md

cline / sdk

cline/cline/sdk/AGENTS.md

Development reference for the Cline SDK workspace.

ClineAGENTS.md

cline / llms

cline/cline/sdk/packages/llms/AGENTS.md

Development guidance for the @cline/llms package.

No other project uses it yet

code-yeongyuAGENTS.md

oh-my-openagent

code-yeongyu/oh-my-openagent/.agents/AGENTS.md

isolated `CODEX_HOME` + local mock model, hook-fired assertions; helper scripts each ship `--self-test` | | `opencode-qa/` | no | opencode CLI/TUI/event-stream QA: hook-fired assertions via SSE, session DB inspection, tmux

code-yeongyuReads credentialsAGENTS.md

oh-my-openagent

code-yeongyu/oh-my-openagent/AGENTS.md

SKIP". THERE IS NO "IT OBVIOUSLY WORKS".** **"It typechecks" is NOT QA. "`bun test` is green" is NOT QA.** YOU MUST DRIVE THE REAL HARNESS, and then **YOU MUST WRITE

code-yeongyuAGENTS.md

oh-my-openagent / assets

code-yeongyu/oh-my-openagent/assets/AGENTS.md

bash bun run build:schema bun run build:omo-schema bun run script/build-help-schemas.ts bun test tests/omo-schema-freshness.test.ts ``` ## CONVENTIONS - Regenerate after changing the corresponding source schema or help definition. - Commit generated output

code-yeongyuAGENTS.md

oh-my-openagent / bin

code-yeongyu/oh-my-openagent/bin/AGENTS.md

selector | | Version mismatch output | `version-mismatch.js` | User-facing installed-wrapper/platform-package mismatch diagnostics | | Behavior tests | `*.test.ts` | Bun tests for invocation routing, platform selection, and diagnostics | ## RUNTIME FLOW `oh-my-opencode.js` derives the invocation

code-yeongyuAGENTS.md

oh-my-openagent / packages

code-yeongyu/oh-my-openagent/packages/AGENTS.md

PACKAGES | Package | Layout | Purpose | |---------|--------|---------| | `lsp-tools-mcp/` | Vendored standalone project (`.github/`, `CHANGELOG.md`, `LICENSE`, `src/`, `test/`, `biome.json`, `vitest.config.ts`) | Serves the 8 aliases `lsp_status`, `lsp_diagnostics`, `lsp_goto_definition`, `lsp_find

code-yeongyuAGENTS.md

oh-my-openagent / scripts

code-yeongyu/oh-my-openagent/scripts/AGENTS.md

pack --dry-run --json` manifests | | `third-party-notice-requirements.mjs` | Declarative notice requirements and path/package matching | | `check-third-party-notices.test.mjs` | Node test coverage for spawn invocation resolution (Windows/non-Windows) and `parseNpmPackJson` (npm 11 array, npm 12 keyed, malformed

code-yeongyuReads credentialsAGENTS.md

oh-my-openagent / tests

code-yeongyu/oh-my-openagent/tests/AGENTS.md

tests/ - Repository-Level Integration Tests ## OVERVIEW Cross-package invariants and integration fixtures that do not belong to one package's co-located unit suite. ## STRUCTURE ``` tests/ ├── omo-config-category-drift.test.ts # Config/category contract stays

docling-projectAGENTS.md

docling

docling-project/docling/AGENTS.md

inside the package (see below) packages/docling/ # full docling meta-package packages/docling-slim/ # slim package readme tests/ # pytest suite and test data docs/ # MkDocs documentation and examples scripts/ # project maintenance scripts ``` ## Skills

No other project uses it yet

remotion-devAGENTS.md

remotion

remotion-dev/remotion/AGENTS.md

Install dependencies (uses Bun) bun install # Build all packages bunx turbo run make # Run tests and linting bunx turbo run lint test # Clean build artifacts bun run clean # Build

No other project uses it yet

openinterpreterAGENTS.md

open-interpreter

openinterpreter/open-interpreter/AGENTS.md

with this fact in mind. It is often used to early exit out of tests that the author knew you would not be able to run given your sandbox limitations