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/bensons/mcp-exec/claude-mdgit clone --depth 1 https://github.com/bensons/mcp-execWhat 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 | $0.01662 | $0.01662 |
| Opus 5 | $0.00831 | $0.00831 |
| Sonnet 5 | $0.00332 | $0.00332 |
| Haiku 4.5 | $0.00166 | $0.00166 |
Grade A, and why
mcp-exec CLAUDE.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 2d 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.
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 — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — mcp-exec
Project-specific working agreements. These are binding for any agent working in this repo.
Handling pull request review feedback
This applies to feedback from human reviewers and from automated reviewers (Codex, Copilot, etc.) alike. Automated reviewers are not authoritative — they are frequently right and occasionally wrong, so every finding gets judged on the evidence, never accepted or dismissed on the strength of who filed it.
For every review thread on a PR you own, do all of the following:
1. Reproduce before you judge
Do not reason about the claim from the diff alone. Run the code and observe the behavior the reviewer describes — typically by comparing the PR build against the merge base:
node -e 'const {Thing} = require("./dist/…"); /* exercise the claim */'
node -e 'const {Thing} = require("/path/to/main/checkout/dist/…"); /* same */'
Capture the actual output. It decides whether you fix, and it is the evidence you quote back to the reviewer. A finding you cannot reproduce is a finding you must not silently "fix".
2. Fix, and push to the PR branch
If the finding is valid, fix it on the same branch and push. Add a regression test that fails without the fix — the test belongs with the fix, in the same commit. Keep the commit message specific about what was reported and what changed.
If fixing the reported problem reveals the same defect one level deeper, fix that too and say so in your reply; do not quietly expand beyond that.
3. Reply in the thread, with details
Reply to the thread itself (not a top-level PR comment) via:
gh api -X POST repos/<owner>/<repo>/pulls/<pr>/comments/<comment_id>/replies -F [email protected]
The reply must state:
- the commit SHA that addresses it,
- the reproduction you ran and its actual output,
- what changed and why that resolves the finding,
- which test now covers it,
- any deliberate deviation from what the reviewer proposed, and the reasoning.
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.
- 2d ago First seen · 156 lines · 1,662 tokens per session scan A 7033b5d909c2
mcp-exec CLAUDE.md is an instructions file published in the GitHub repository bensons/mcp-exec (0 stars, last pushed 3d ago), licensed MIT. It adds 1,662 tokens to every session, about $0.0083 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 2026-08-31.
Other instructions, from other repositories
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).
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.
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.
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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.