AgentLint: Instructions file for Claude Code

CLAUDE.md

AgentLint CLAUDE.md is an instructions file for Claude Code from 0xmariowu/AgentLint. It costs 653 tokens per session, scanned A, original, MIT.

A set of repository instructions for AgentLint, including its rules, required local tests, evidence standards, and workflow. AgentLint is a tool for checking how well a repository supports AI coding agents.

In plain words
What is it for?
Use it when modifying AgentLint checks, scoring, reports, plans, or scanner output, and when running its compliance and scanner tests before pushing changes.
Why use it?
It keeps changes to the checker grounded in documented evidence rather than personal opinion. It also explains which related files must change when the scanner’s output or scoring changes.

Instructions file for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: mentions CLAUDE.md.

This is 0xmariowu/AgentLint's own configuration. It tells Claude Code how to work on AgentLint itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything AgentLint configures →

Runs only inside a plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else, and the catalogue could not identify which plugin ships it.

Reuse

Borrowing it

Nothing to install: this file belongs to 0xmariowu/AgentLint. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/0xmariowu/AgentLint/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/0xmariowu/AgentLint

Made for: Claude Code.

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 AgentLint CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/0xmariowu/agentlint/claude-md.svg)](https://agentmods.dev/instructions/0xmariowu/agentlint/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/0xmariowu/agentlint/claude-md"><img src="https://agentmods.dev/badge/instructions/0xmariowu/agentlint/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 653 This file is loaded in full into every session.
When invoked 653 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00653 $0.00653
Opus 5 $0.00327 $0.00327
Sonnet 5 $0.00131 $0.00131
Haiku 4.5 $0.00065 $0.00065

Measured 8d ago against content hash 8ffef93b92f6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

AgentLint 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 8d 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 · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentLint

For global rules, see ~/CLAUDE.md. This file contains agent-lint-specific overrides only. Lighthouse for AI-Friendly repos — diagnose, plan, fix.

Local test (run before push)

bash tests/test_compliance.sh && bash tests/test-scanner.sh

Compliance (4 checks, ~1s) + scanner unit tests. Full matrix (multi-OS) runs in CI.

Session Checklist

  1. If modifying checks → read standards/evidence.json for evidence backing
  2. If modifying scoring → read standards/weights.json for weight rationale
  3. If adding a check → must have evidence source (paper, data, or real audit finding)

Rules

  • Don't add a check without empirical evidence. Instead, cite: Anthropic data, academic paper, or documented real-world failure. Because: AgentLint's value is evidence-backed recommendations, not opinions.
  • Don't hardcode thresholds as pass/fail gates. Instead, measure and compare to reference values. Because: Bitter Lesson — imposing human-defined boundaries creates brittle rules that break when context changes.
  • Don't modify scanner output format without updating scorer + plan-generator + reporter. Because: the JSONL schema is a contract between all pipeline stages.
  • Don't hardcode paths in skills/*.md. Instead, use ${CLAUDE_PLUGIN_ROOT} to reference bundled files. Because: plugin install path varies per user.
  • Don't write persistent data to ${CLAUDE_PLUGIN_ROOT}. Instead, use ${CLAUDE_PLUGIN_DATA} or ~/.al/. Because: plugin root is replaced on update.
  • When you change a file under .github/workflows/, also check whether templates/universal/<basename> exists and needs the same change. Instead of relying on memory, run bash scripts/lib/check-template-sync.sh against your staged diff (the pre-commit hook already does this informally) — it counts security-keyword occurrences in both files and warns when they drift. Because: v1.1.9 / v1.1.10 / v1.1.11 each shipped a regression where the in-tree workflow got a security fix but the user-facing template did not, exposing every fresh agentlint setup repo to the same bug.

Read the full file on GitHub · 46 lines

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. 8d ago First seen · 46 lines · 653 tokens per session scan A 8ffef93b92f6

Subscribe to this mod's changes

AgentLint CLAUDE.md is an instructions file published in the GitHub repository 0xmariowu/AgentLint (55 stars, last pushed 1mo ago), licensed MIT. It adds 653 tokens to every session, about $0.0033 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.

Related

Other instructions, from other repositories