How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
OpenAIAGENTS.mdopenai/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
LangChainAGENTS.mdlangchain-ai/open-swe/AGENTS.md
controls may ship without a corresponding tool, especially for secret input until a secure tool-driven input flow exists. Do not add a tool when the agent can already
GoogleAGENTS.mdgoogle/adk-recipes/core/python/long-horizon-harness/AGENTS.md
prefix, prefix cache, the four bounds on context growth), and [`docs/security-model.md`](docs/security-model.md) (per-layer security). Don't duplicate those here — link to them. ## What this recipe teaches Horizon leans
KuberwastakenAGENTS.mdKuberwastaken/claurst/docs/agents.md
same name. ```json { "agents": { "review": { "description": "Senior code reviewer focused on correctness and security", "model": "anthropic/claude-opus-4-6", "temperature": 0.3, "prompt": "You are a senior software engineer performing code review. Focus
CloudflareAGENTS.mdcloudflare/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.mdcloudflare/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
An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.
At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.
Setup and test commands, code style, and the rules a new contributor would need to know.
Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.
