Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/skyvern-ai/rustwright/agents-mdgit clone --depth 1 https://github.com/Skyvern-AI/rustwrightWhat 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 | $0.00996 | $0.00996 |
| Opus 5 | $0.00498 | $0.00498 |
| Sonnet 5 | $0.00199 | $0.00199 |
| Haiku 4.5 | $0.00100 | $0.00100 |
Grade A, and why
rustwright AGENTS.md scanned grade A with 0 findings 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 yesterday.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Instructions
Public boundary
Treat every tracked file, commit, branch name, and review artifact as if it will be publicly visible. This repository is the source of truth; a separate, reviewed sync republishes approved content to a public mirror, so every change merged here is the last gate before it can become public.
- Never add credentials, auth headers, customer data, private URLs, personal filesystem paths, employee email addresses, internal ticket identifiers, or private host/network topology.
- Do not name private repositories, packages, modules, services, or operational
incidents. Use neutral placeholders such as
<organization>,<repository>,<remote-host>, and<source-checkout>in public examples. - Keep integration tools that import or inspect private code in the private repository. Public tools must require explicit user-supplied paths and must not default to sibling checkouts.
- Keep raw benchmark data, terminal output, HAR files, environment files, and generated reports untracked. Publish only reviewed, reproducible summaries; never let a headline number rest on demo-grade or otherwise irreproducible data.
- Use a GitHub
noreplyaddress for commit author and committer identities. Automation co-author trailers must likewise use a providernoreplyaddress (for example[email protected]), never a personal or corporate mailbox. Never rewrite or force-push shared history without explicit owner approval, and never usegit push --all.
Before committing, inspect the staged diff for disclosure risks.
Pull-request review
Development checkouts push only to this repository — there is no direct push to
the public mirror; the reviewed sync is the only path content takes to become
public. Every pull request opened here is reviewed automatically by
.github/workflows/claude-code-review.yml, which loads its prompt from
.github/prompts/code-review.md on the pull request's base branch (so an
untrusted pull request cannot alter its own review) and posts a single comment.
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.
- yesterday First seen · 89 lines · 996 tokens per session scan A f5833d4d77a0
rustwright AGENTS.md is an instructions file published in the GitHub repository Skyvern-AI/rustwright (863 stars, last pushed 6d ago), licensed MIT. It adds 996 tokens to every session, about $0.0050 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
cdpilot CLAUDE.md
Instructions for mehmetnadir/cdpilot, covering cdpilot, kimlik, hızlı başlangıç, dosya haritası and dikkat edilecekler.
cdpilot GEMINI.md
Instructions for mehmetnadir/cdpilot, covering context-mode — mandatory routing rules, blocked commands — do not attempt these, curl / wget — blocked, inline http — blocked and webfetch / web browsing — blocked.
eyebrowse CLAUDE.md
Claude Code instructions for Evil-Bane/eyebrowse, covering claude.md — eyebrowse build notes, layout, run, pinned versions (lockstep matters) and key constraints (encoded in code — don't "simplify" away).
chrome-cdp-ex CLAUDE.md
Instructions for EndeavorYen/chrome-cdp-ex, covering claude.md — chrome-cdp-ex, problem-solving principles, project overview, architecture and coding conventions.
cdpilot AGENTS.md
Instructions for mehmetnadir/cdpilot, covering gitnexus — code intelligence, always do, when debugging, when refactoring and never do.
gaze AGENTS.md
AGENTS.md instructions for KevinTrinhDev/gaze, covering agents.md - gaze, if you are an ai agent, read this before you run anything, the short version, before you debug something and traps that repeat.