AGENTS.md examples from real projects

dyad-shAGENTS.md

dyad

dyad-sh/dyad/AGENTS.md

will likely give wrong results. ## Project context - This is an Electron application with a secure IPC boundary. - Frontend is a React app that uses TanStack Router (not Next.js or React

No other project uses it yet

HomebrewAGENTS.md

homebrew-cask

Homebrew/homebrew-cask/AGENTS.md

stars) - No public presence beyond "just `brew install`" - Unmaintained or has unpatched security vulnerabilities - Requires SIP to be disabled - Download requires login/registration (walled builds) - No compiled versions available (use homebrew/core

MobSFAGENTS.md

Mobile-Security-Framework-MobSF

MobSF/Mobile-Security-Framework-MobSF/AGENTS.md

MobSF Agent Guidelines MobSF is a security analysis platform. Every code path processes attacker-supplied input (APKs, ZIPs, IPAs, manifests) from authenticated but potentially malicious users. Security must

No other project uses it yet

handsontableReads credentialsAGENTS.md

handsontable

handsontable/handsontable/AGENTS.md

dependencies. - Adding dependencies under non-permissive licenses requires notifying clients through the Sales team. --- ## Security - Strict XSS prevention in user-facing cell content. - Input sanitization on custom formulas and cell

handsontableAGENTS.md

handsontable / handsontable

handsontable/handsontable/handsontable/AGENTS.md

both lie when the grid renders inside a shadow root, and Salesforce Lightning Web Security additionally filters `composedPath()` down to the host chain and collapses `activeElement` to the outermost shadow

QwikDevAGENTS.md

qwik

QwikDev/qwik/AGENTS.md

unit.ts(x)` | Vitest unit files | | `*.spec.ts(x)` | Vitest spec files | | `*.e2e.ts` | Playwright e2e files | ### Security And Supply Chain Treat security-sensitive changes as behavior changes even when they look like

QwikDevAGENTS.md

qwik

QwikDev/qwik/.ruler/AGENTS.md

unit.ts(x)` | Vitest unit files | | `*.spec.ts(x)` | Vitest spec files | | `*.e2e.ts` | Playwright e2e files | ### Security And Supply Chain Treat security-sensitive changes as behavior changes even when they look like

matomo-orgAGENTS.md

matomo

matomo-org/matomo/AGENTS.md

range review | `matomo-review` | | In-development debt or maintainability review | `matomo-debt-check` | | Security-sensitive changes: access control, CSRF, SQL, request parsing, tokens, secrets | `matomo-security-rules` | | Public API methods

NirDiamantAGENTS.md

agents-towards-production

NirDiamant/agents-towards-production/AGENTS.md

independent tutorial tracks, each covering one layer of the production stack: orchestration, memory, retrieval, security and guardrails, observability and tracing, evaluation, deployment, GPU serving, fine-tuning and user interfaces

netbox-communityReads credentialsAGENTS.md

netbox

netbox-community/netbox/AGENTS.md

ruff) and the full test suite across the supported Python version matrix. - **`codeql.yml`** — CodeQL security scanning. - **`claude.yml`** — Claude Code automation hook; triggers on issue/PR comments mentioning `@claude`. - **`claude-issue-triage.yml`** — Automated issue

oracleAGENTS.md

graal

oracle/graal/AGENTS.md

included when behavior changes. - [ ] Flag and ask for extra permission for changes related to security (e.g., touching cryptographic code

NVIDIAReads credentialsAGENTS.md

OpenShell

NVIDIA/OpenShell/AGENTS.md

OCSF builder + `ocsf_emit!()` for events that represent **observable sandbox behavior** visible to operators, security teams, or agents monitoring the sandbox: - Network decisions (allow, deny, bypass detection) - HTTP/L7 enforcement decisions

containerdAGENTS.md

containerd

containerd/containerd/AGENTS.md

docs/garbage-collection.md) | | Namespace propagation through context | [Namespaces](docs/namespaces.md) | | Daemon-side image transfers | [Transfer service](docs/transfer.md) | ## Security findings Before scanning for security issues, evaluating a suspected vulnerability, or drafting a security report

GitHubAGENTS.md

docs / responsible-use

github/docs/content/copilot/responsible-use/agents.md

assigned issues. The cloud agent runs in an ephemeral, firewalled environment with automated security scanning. * **Copilot Cloud**: Copilot in the Cloud (GitHub Copilot on GitHub.com, GitHub Mobile, or anywhere that

xbmcAGENTS.md

xbmc

xbmc/xbmc/AGENTS.md

issues that should block the change. * Report only regressions, correctness problems, build failures, security issues, or behavior that contradicts the stated intent. * Do not suggest optional improvements, cleanup, alternative wording

vulhubAGENTS.md

vulhub

vulhub/vulhub/AGENTS.md

Guide Vulhub is an open-source collection of pre-built vulnerable Docker environments for security education. ## Project Structure ``` base/ / /Dockerfile # Base Docker images for vulnerable software / / # Vulnerability environments ├── docker-compose.yml ├── README.md

No other project uses it yet

vitessioAGENTS.md

vitess / changelog

vitessio/vitess/changelog/AGENTS.md

fixes, such as stricter validation that rejects input Vitess used to accept - Security fixes that have a published advisory. Until the advisory is public, leave

PydanticReads credentialsAGENTS.md

pydantic-ai / workflows

pydantic/pydantic-ai/.github/workflows/AGENTS.md

green while never having run. That is not hypothetical: the former `pydantic-ai-ui-security-review` shipped this way and its agent never fired once, reporting false-green on every

PydanticReads credentialsAGENTS.md

pydantic-ai / realtime

pydantic/pydantic-ai/pydantic_ai_slim/pydantic_ai/realtime/AGENTS.md

reimplement or filter above it. A reimplemented policy layer that drifts is a security bug, not a style problem: the realtime approval-bypass was exactly this. - When behavior differs from

wekanAGENTS.md

wekan

wekan/wekan/AGENTS.md

pull request ITSELF is the participant, and is named.** GitHub CodeQL filing a security fix, Copilot Autofix, Dependabot raising a dependency bump: nobody wrote those, so crediting a human would