undercover CLAUDE.md

undercover CLAUDE.md is an instructions file for coding agents from anasfik/undercover. It costs 225 tokens per session, scanned A, original, MIT.

An instruction file for a tool that removes signs of AI-generated writing from code comments and Git commit messages before public sharing.

In plain words
What is it for?
Use it before pushing to a public repository, opening a pull request, or reviewing recently edited files for AI-like comment patterns.
Why use it?
It helps prepare open-source contributions without changing the program's logic, text strings, or variable names.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/anasfik/undercover/claude-md
Clone the repo
git clone --depth 1 https://github.com/anasfik/undercover

Wrote 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.

agentmods badge for undercover CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/anasfik/undercover/claude-md.svg)](https://agentmods.dev/instructions/anasfik/undercover/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/anasfik/undercover/claude-md"><img src="https://agentmods.dev/badge/instructions/anasfik/undercover/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 225 This file is loaded in full into every session.
When invoked 225 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00225 $0.00225
Opus 5 $0.00112 $0.00112
Sonnet 5 $0.00045 $0.00045
Haiku 4.5 $0.00022 $0.00022

Measured 3d ago against content hash f35dee7ae2d0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

undercover 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 3d 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.

CLAUDE.md · 23 lines

What it actually says

Undercover Mode

This repository provides undercover mode — an AI agent skill that strips AI-characteristic patterns from code and commit messages before they are pushed to public repositories.

Available skill

/undercover [staged|HEAD|HEAD~N|<file>] [--dry-run]

Use this skill when preparing a contribution for an open-source project. It scans for AI-labeling fingerprints, shows you exactly what it found, asks for confirmation, then cleans the files and optionally fixes the commit message.

When to use it (without being asked)

Automatically suggest running /undercover when:

  • The user is about to run git push on a branch targeting a public repository
  • You detect AI-characteristic comment patterns in files the user just edited
  • The user asks to "clean up before the PR" or "prep this for open source"

What the skill touches

Only comment lines and commit messages. Never logic, strings, or variable names.

See .claude/skills/undercover/SKILL.md for the full pattern list and execution protocol.

Changes

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.

  1. 3d ago First seen · 23 lines · 225 tokens per session scan A f35dee7ae2d0

Subscribe to this mod's changes

undercover CLAUDE.md is an instructions file published in the GitHub repository anasfik/undercover (2 stars, last pushed 5mo ago), licensed MIT. It adds 225 tokens to every session, about $0.0011 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.