Borrowing it
Nothing to install: this file belongs to xChechi/xche-ai-app-security-pack. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/xChechi/xche-ai-app-security-pack/main/AGENTS.mdgit clone --depth 1 https://github.com/xChechi/xche-ai-app-security-packWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/xchechi/xche-ai-app-security-pack/agents-md)<a href="https://agentmods.dev/instructions/xchechi/xche-ai-app-security-pack/agents-md"><img src="https://agentmods.dev/badge/instructions/xchechi/xche-ai-app-security-pack/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00763 | $0.00763 |
| Opus 5 | $0.00381 | $0.00381 |
| Sonnet 5 | $0.00153 | $0.00153 |
| Haiku 4.5 | $0.00076 | $0.00076 |
Grade B, and why
xche-ai-app-security-pack AGENTS.md scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Cloud metadata endpointmediumServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
- **SSRF:** never let user input control a server-side request target; allow-list destinations; block private/link-local ranges incl. `169.254.169.254`. Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
What it actually says
AGENTS.md — Security rules (cross-tool)
The portable, cross-tool version of these rules. Read natively by Codex, Cursor, GitHub Copilot, Gemini CLI, Aider, Windsurf, Zed and 25+ other agents. Claude Code users:
CLAUDE.mdcarries the same rules plus Claude-specific extras. Full policy with rationale: seeSECURITY.md.These are defaults. If a task conflicts with a rule, stop and flag it — never bypass a control to "make it work."
- Never hardcode secrets. Read keys/tokens/DB URLs from env vars. Ensure
.env/*.env*is gitignored before the first commit. Never put secrets/PII in client code, logs, URLs, or error messages. - SQL: always parameterized / ORM. Never build queries with string concatenation, f-strings,
.format(),%, or template literals containing user input. - XSS: never feed user input to
dangerouslySetInnerHTML,innerHTML,v-html,|safe, ordocument.write. Rely on framework escaping; encode output for its context. - No command injection: never pass input to a shell,
eval,new Function, or unsafe deserialization (pickle, unsafeyaml.load). - Validate every input server-side against a schema (Zod/Pydantic), allow-list not block-list. Client-side checks are UX only.
- Access control: every endpoint that reads/writes an object must verify ownership/permission server-side. No exceptions for "internal" or unlinked admin routes.
- Multi-tenant: scope every query to the caller's tenant. Cross-tenant access (IDOR/BOLA) is a critical bug — assume it's there until proven otherwise.
- Mass assignment: bind request bodies to explicit allow-listed fields; never let users set their own role/owner/price/tenant.
- Passwords: hash with argon2id or bcrypt — never MD5/SHA-1/unsalted. Rate-limit auth endpoints; generic "invalid credentials" messages.
- Sessions/cookies:
HttpOnly+Secure+SameSite. Never store secrets/sensitive data in a readable signed cookie or JWT payload. - Security headers: set CSP (Report-Only first), HSTS,
X-Frame-Options: DENY,X-Content-Type-Options: nosniff,Referrer-Policy. NoAccess-Control-Allow-Origin: *on authenticated endpoints. - SSRF: never let user input control a server-side request target; allow-list destinations; block private/link-local ranges incl.
169.254.169.254. - File uploads: validate by content type, cap size, store outside web root with generated names, never execute.
- Dependencies: verify a package actually exists before adding it (no hallucinated/slopsquatted names); pin versions; run the audit tool; keep deps minimal.
- LLM/MCP: treat model + tool output as untrusted (encode before render/execute); sandbox MCP servers, allow-list commands by absolute path; require human confirmation for high-impact actions; never send secrets or unminimized PII to models.
- Errors: return generic messages to clients; never leak stack traces, SQL, or internal paths. Log security events but never secrets/PII.
- Process: for changes touching auth, payments, secrets, or LLM tools, say so explicitly. Never weaken an existing control to pass a test. Confirm fixes on staging, not prod.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 28 lines · 763 tokens per session scan B b0ea9b2ec082
xche-ai-app-security-pack AGENTS.md is an instructions file published in the GitHub repository xChechi/xche-ai-app-security-pack (5 stars, last pushed 2mo ago), licensed MIT. It adds 763 tokens to every session, about $0.0038 per session on Opus 5. A static security scan graded it B with 1 finding (cloud metadata endpoint). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.