AGENTS.md examples from real projects

GoogleAGENTS.md

gvisor

google/gvisor/AGENTS.md

assumptions by writing syscall tests to verify Linux behavior from a black-box perspective. #### Security first All changes are reviewed for security. gVisor's design and architecture ensures that

sandboxie-plusAGENTS.md

Sandboxie

sandboxie-plus/Sandboxie/AGENTS.md

change guidelines - Treat `Sandboxie/core` with extreme caution: bugs can cause BSODs or sandbox-escape security holes. Prefer minimal, surgical patches there over refactors. - Verify builds, branches, and properties

agent0aiAGENTS.md

agent-zero / agents

agent0ai/agent-zero/agents/AGENTS.md

metadata and inherited prompt specifics. | | [developer/AGENTS.md](developer/AGENTS.md) | Software development specialist profile. | | [hacker/AGENTS.md](hacker/AGENTS.md) | Cyber security and penetration testing specialist profile. | | [researcher/AGENTS.md](researcher/AGENTS.md) | Research, data analysis, and reporting specialist profile. | | [tiny

No other project uses it yet

agent0aiAGENTS.md

agent-zero / api

agent0ai/agent-zero/api/AGENTS.md

changing handler behavior. - For auth, CSRF, upload/download, tunnel, or file endpoints, run the nearest security regression tests. - Check file-level documentation coverage with a script or shell loop that verifies

No other project uses it yet

agent0aiAGENTS.md

agent-zero / extensions

agent0ai/agent-zero/extensions/AGENTS.md

repeatedly when the lifecycle point can fire multiple times. - Secret masking, auth, security, and persistence extensions must not be bypassed by convenience changes. ## Work Guidance - Keep extension code small

No other project uses it yet

agent0aiAGENTS.md

agent-zero / helpers

agent0ai/agent-zero/helpers/AGENTS.md

belong in `api/`, tool behavior in `tools/`, and plugin-local logic inside plugin directories. - Security, auth, settings, file access, plugin discovery, extension dispatch, notifications, state snapshots, scheduler, tunnel, and WebSocket

No other project uses it yet

agent0aiAGENTS.md

agent-zero / plugins

agent0ai/agent-zero/plugins/AGENTS.md

update flows from ZIP, Git, and Plugin Index. | | [_plugin_scan/AGENTS.md](_plugin_scan/AGENTS.md) | LLM-guided security scanner for third-party plugins. | | [_plugin_validator/AGENTS.md](_plugin_validator/AGENTS.md) | Plugin manifest, structure, convention, and security

No other project uses it yet

agent0aiAGENTS.md

agent-zero / references

agent0ai/agent-zero/skills/a0-create-plugin/references/AGENTS.md

handlers. - Keep local-only work separate from community publication and catalog discovery separate from security scanning. - Preserve authentication, scope, user authorization, and unrelated runtime state. ## Work Guidance - Prefer existing framework

No other project uses it yet

agent0aiAGENTS.md

agent-zero / a0-manage-plugin

agent0ai/agent-zero/skills/a0-manage-plugin/AGENTS.md

behavior, scan expectations, and activation semantics current. - Distinguish Index discovery from model-based security scans and installation; metadata is not verified capability or safety. - Handle missing/null Index fields and check

agent0aiAGENTS.md

agent-zero / references

agent0ai/agent-zero/skills/a0-manage-plugin/references/AGENTS.md

handlers. - Keep local-only work separate from community publication and catalog discovery separate from security scanning. - Preserve authentication, scope, user authorization, and unrelated runtime state. ## Work Guidance - Prefer existing framework

No other project uses it yet

agent0aiAGENTS.md

agent-zero / tests

agent0ai/agent-zero/tests/AGENTS.md

Tests DOX ## Purpose - Own pytest regression, security, integration, and contract tests. - Keep tests focused on behavior that should remain stable across framework changes. ## Ownership - Test files live directly under `tests

No other project uses it yet

elizaOSAGENTS.md

eliza

elizaOS/eliza/AGENTS.md

desktop/mobile inspection. - Submit changes through a PR against `develop`. Report vulnerabilities privately through GitHub Security Advisories. The community plugin registry is retired; first-party catalog data lives in `packages/core/src/catalog/`. Getting

No other project uses it yet

browser-useAGENTS.md

browser-harness

browser-use/browser-harness/AGENTS.md

routine PR gates. - First-time install / blocked Chrome: follow `install.md` (`chrome://inspect/#remote-debugging`). # Security - Do not commit secrets, Browser Use Cloud tokens, or session cookies. - Prefer the smallest change

No other project uses it yet

facebookAGENTS.md

yoga

facebook/yoga/AGENTS.md

just that platform. Choose the appropriate type: `[Breaking]`, `[Added]`, `[Changed]`, `[Deprecated]`, `[Removed]`, `[Fixed]`, or `[Security]`. If the change truly has no user-visible effect, such as a test-only

lidge-junPipes a download into a shellAGENTS.md

opencodex

lidge-jun/opencodex/AGENTS.md

gitlink may be tracked anywhere, and neither the vendored reference clones nor the security triage excised before publication may reappear in the index. Both were driven red once to prove

lidge-junAGENTS.md

opencodex

lidge-jun/opencodex/.github/AGENTS.md

instructions This file applies to `.github/` and inherits the repository-wide rules in `/AGENTS.md`. ## Security boundary Every workflow, ownership, branch-enforcement, release, or repository-automation change requires explicit security review

lidge-junAGENTS.md

opencodex / gui

lidge-jun/opencodex/gui/AGENTS.md

provided by the current stack or a small local implementation. - Dependency changes require explicit security review. - Update `docs-site/` when dashboard behavior, setup, or configuration changes for users. ## Failure mode

lidge-junAGENTS.md

opencodex / scripts

lidge-jun/opencodex/scripts/AGENTS.md

Changes to release, packaging, dependency-installation, credential, or repository-automation scripts require explicit security review. ## Implementation rules - Preserve Linux, macOS, and Windows behavior. Do not add shell-specific assumptions

lidge-junAGENTS.md

opencodex / src

lidge-jun/opencodex/src/AGENTS.md

relevant to that adapter. - Authentication, OAuth, token, credential, management API, and CORS changes are security-boundary changes. ## Tests and validation - Place focused regression coverage near the existing tests

lidge-junAGENTS.md

opencodex / structure

lidge-jun/opencodex/structure/AGENTS.md

devlog/`. - Superseded or alternative reasoning belongs in `decisions/`, not in the doc body. - Unreleased security findings belong in scratch space and nowhere in this repository. The rule in the root