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/eslamx7/ai-trailers/claude-mdgit clone --depth 1 https://github.com/EslaMx7/ai-trailersWrote 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/eslamx7/ai-trailers/claude-md)<a href="https://agentmods.dev/instructions/eslamx7/ai-trailers/claude-md"><img src="https://agentmods.dev/badge/instructions/eslamx7/ai-trailers/claude-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 | $0.00555 | $0.00555 |
| Opus 5 | $0.00278 | $0.00278 |
| Sonnet 5 | $0.00111 | $0.00111 |
| Haiku 4.5 | $0.00056 | $0.00056 |
Grade A, and why
ai-trailers 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 5d 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project
ai-trailers is a Bun-only CLI tool (bunx ai-trailers) that captures AI coding tool prompts and embeds them as standard git trailers in commit messages. No runtime dependencies — uses only Bun built-ins and Node.js standard library.
Commands
bun ./src/cli.ts <command> # Run CLI directly during development
bun install # Install dev dependencies
bun test # Run tests (bun:test)
Bun conventions
Default to Bun instead of Node.js for everything: bun not node, bunx not npx, bun install not npm install. Prefer Bun.file/Bun.write over node:fs readFile/writeFile. Bun auto-loads .env — don't use dotenv.
Architecture
Data flow: AI Tool hook → capture command → .ai-trailers file → commit-msg git hook → commit message trailers → file cleared.
Modules:
src/cli.ts— Entry point and command dispatcher. Imports all other modules. Reads version frompackage.json.src/tools.ts— Tool registry. Single source of truth for supported tools (name, marker files, hook event, config format). To add a new tool, add one entry to thetoolsarray.src/detect.ts— Detects AI tools by checking for marker files/directories. Uses the registry fromtools.ts.src/extractors.ts— Reusable prompt extraction helpers:extractFromStdin({ format, path })for JSON/text stdin,extractFromEnv(varName)for environment variables. Each tool defines which extractor to use.src/capture.ts— Looks up the tool by name, calls itsextractPrompt(), appends formatted git trailers to.ai-trailers. Called asai-trailers capture --tool <name>.src/install.ts— Installs/removes tool hook configs (deep-merges into existing settings files), thecommit-msggit hook, and.gitignoreentries. Respectscore.hooksPath.
Key design: Each tool defines its own extractPrompt function in tools.ts using helpers from extractors.ts. Tools differ in how they pass the prompt (stdin JSON, env var) and their hook config format. Adding a new tool is one entry in the tools array.
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.
- 5d ago First seen · 35 lines · 555 tokens per session scan A 1c5197aac300
ai-trailers CLAUDE.md is an instructions file published in the GitHub repository EslaMx7/ai-trailers (21 stars, last pushed 5mo ago), licensed MIT. It adds 555 tokens to every session, about $0.0028 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
gemini-rules-repo GEMINI.md
Instructions for sanpicule/gemini-rules-repo, covering gemini 開発ルール, 1. コミットメッセージ, 2. セキュリティ, 3. ブランチモデル and ブランチの種類.
undercover CLAUDE.md
Instructions for anasfik/undercover, covering undercover mode, available skill, when to use it (without being asked) and what the skill touches.
TLDR GEMINI.md
Instructions for 0p9b/TLDR: @./skills/tldr/SKILL.md @./skills/tldr-commit/SKILL.md @./skills/tldr-review/SKILL.md @./skills/tldr-compress/SKILL.md.
undercover copilot-instructions.md
Instructions for anasfik/undercover, covering undercover mode, trigger phrases, what to strip from source files, what to fix in commit messages and protocol.
planning-with-files AGENTS.md
AGENTS.md instructions for OthmanAdi/planning-with-files, covering agents.md — planning-with-files agent reference card, commit rules, release checklist (12 steps), version bump scope and changelog format.
ClawRouter CLAUDE.md
Claude Code instructions for BlockRunAI/ClawRouter, covering clawrouter, commands, project structure, key dependencies and conventions.