AGENTS.md examples from real projects

BeehiveInnovationsReads credentialsAGENTS.md

zen-mcp-server

BeehiveInnovations/zen-mcp-server/AGENTS.md

Install GitHub CLI: `brew install gh` (macOS) or visit https://cli.github.com for other platforms. ## Security & Configuration Tips Store API keys and provider URLs in `.env` or your MCP client config

pestphpReads credentialsAGENTS.md

pest

pestphp/pest/AGENTS.md

that shows a secret, such as `env`, `printenv`, `set`, `cat .env`, `gh auth token`, `security find-generic-password` or `git config --get-regexp credential`. When you need only the names

betaflightAGENTS.md

betaflight

betaflight/betaflight/AGENTS.md

give a reason in one line. - *No value*: give the reason only. 3. **Security and data loss.** Do not give a security or data-loss finding the level *No value

pi-holeAGENTS.md

docker-pi-hole

pi-hole/docker-pi-hole/AGENTS.md

many systems. - The correct project spelling is "Pi-hole" (capital P, lowercase h, hyphen). ## Security considerations - Secrets can be supplied via Docker secrets as well as environment variables; never

No other project uses it yet

grafanaAGENTS.md

pyroscope

grafana/pyroscope/AGENTS.md

commit changes to `node_modules/` or generated code without source changes ## Security Considerations 1. **Input Validation**: Always validate and sanitize user input 2. **Path Traversal**: Validate object keys before storage

No other project uses it yet

CloudflareReads credentialsAGENTS.md

cloudflare-os

cloudflare/cloudflare-os/AGENTS.md

rethink the design); and prefer reusing existing mechanisms over adding parallel ones. Capability-based security note: a resource becomes "ambient" (auto-injected) only by user/admin configuration — a gatekeeper must never

CloudflareAGENTS.md

cloudflare-os / configurator-ui

cloudflare/cloudflare-os/packages/configurator-ui/AGENTS.md

HTML document: the minified runtime, its Tailwind/Kumo CSS, and the module, under a content security policy that admits only those scripts by hash and allows no network access. - Kumo

CloudflareAGENTS.md

cloudflare-os / workshop-frontend

cloudflare/cloudflare-os/packages/workshop-frontend/AGENTS.md

block already says. Comments are appropriate for gotchas, non-obvious invariants, external constraints, security or performance reasoning, and deliberate departures from a convention. Explain why the surprising choice is necessary

WhiskeySocketsReads credentialsAGENTS.md

Baileys

WhiskeySockets/Baileys/AGENTS.md

separately; bundle them in their own PR when possible. - **`package.json` `resolutions`** — these patch known security advisories. Removing entries reintroduces CVEs; check `yarn npm audit --recursive` before pruning. ## Testing expectations

No other project uses it yet

apereoAGENTS.md

cas

apereo/cas/AGENTS.md

skill fits. | Skill | Read when the work touches | |---|---| | [`cas-actuators`](.claude/skills/cas-actuators/SKILL.md) | Actuator endpoints, management security | | [`cas-attribute-consent`](.claude/skills/cas-attribute-consent/SKILL.md) | `support/cas-server-support-consent*` | | [`cas-build-ci`](.claude/skills/cas-build-ci/SKILL.md) | Gradle build logic, `.github/workflows`, `ci/` scripts

spinabotReads credentialsAGENTS.md

brigade

spinabot/brigade/AGENTS.md

filesystem) + convex/ ├── config/ # brigade.json schema, io, validators ├── providers/ # model provider catalog + auth detection ├── auth/ · security/ # auth profiles, encryption (libsodium seal) ├── cron/ # scheduler + isolated-agent run executor ├── tui/ · ui/ # terminal client

QuipNetworkAGENTS.md

ethereum-sdk

QuipNetwork/ethereum-sdk/AGENTS.md

WalletFactory` keeps `_authorizeUpgrade` with its other internals near the file end. Rationale: on these security contracts, adjacency of each override and its helpers keeps the code auditable and preserves

OpenAIAGENTS.md

codex-security

openai/codex-security/AGENTS.md

Keep it simple Codex Security is a thin wrapper around Codex and its security plugin. - Trust local tools and processes running as the current user. - Treat repository contents, model output

No other project uses it yet

OpenAIAGENTS.md

codex-security / typescript

openai/codex-security/sdk/typescript/AGENTS.md

Codex Security CLI Codex Security is a thin CLI and SDK wrapper around Codex and its security plugin. Use their existing behavior instead of building another runtime. ## Threat model

No other project uses it yet

Model Context ProtocolReads credentialsAGENTS.md

inspector

modelcontextprotocol/inspector/AGENTS.md

merge, release notes (incl. reporter thanks), tag `origin/main`, publish, re-cut on failure | `/release` | | [`security-advisory`](.claude/skills/security-advisory/SKILL.md) | A privately reported vulnerability end to end: the draft card, who owns

OpenAIAGENTS.md

openai-node

openai/openai-node/AGENTS.md

meaningful public JSDoc. Preserve documentation URLs/import paths; examples must run with documented environments/dependencies. ## Security and lifecycle correctness Read and apply [contributor security requirements](.github/CONTRIBUTING.md#security-requirements) for credentials/synthetic fixtures, browser

No other project uses it yet

TEN-frameworkAGENTS.md

ten-framework

TEN-framework/ten-framework/AGENTS.md

them upfront. This gives you setup, architecture, code map, conventions, workflows, interfaces, gotchas, and security. 3. If a task needs more detail than L1 provides, follow links to L2 deep

bunkerityAGENTS.md

bunkerweb

bunkerity/bunkerweb/AGENTS.md

changes with maintainers and flag them clearly in both commit body and PR description. ## Security & Configuration Tips Never commit secrets—use sample files in `env/` or add new templates when

No other project uses it yet

tambo-aiAGENTS.md

tambo

tambo-ai/tambo/.github/AGENTS.md

GitHub Actions workflows: - **Pin actions to full commit SHA** with a version comment for security and reproducibility: ```yaml # Good - pinned to SHA with version comment - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2

LangChainAGENTS.md

open-swe

langchain-ai/open-swe/AGENTS.md

tests that do. - No speculative edge cases or the same behavior tested across layers. Security, authorization, data integrity, and tricky state transitions do deserve targeted coverage. - For bug fixes, prefer

No other project uses it yet