ruleProb CLAUDE.md

ruleProb CLAUDE.md is an instructions file for Claude Code from canblmz1/ruleProb. It costs 2,010 tokens per session, scanned A, original, MIT.

A project guide for RuleProbe, a command-line tool that extracts rules from files such as CLAUDE.md and tests whether AI agents follow them.

In plain words
What is it for?
Finding core files, generating test scenarios, running agent providers, comparing results, optimising rules, and producing reports.
Why use it?
It gives agents a map of the project, its main code areas, and the commands used to validate repository rules.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions AGENTS.md.

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/canblmz1/ruleprob/claude-md
Clone the repo
git clone --depth 1 https://github.com/canblmz1/ruleProb

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 ruleProb CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/canblmz1/ruleprob/claude-md.svg)](https://agentmods.dev/instructions/canblmz1/ruleprob/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/canblmz1/ruleprob/claude-md"><img src="https://agentmods.dev/badge/instructions/canblmz1/ruleprob/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,010 This file is loaded in full into every session.
When invoked 2,010 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 5.1 $0.02010 $0.02010
Opus 5 $0.01005 $0.01005
Sonnet 5 $0.00402 $0.00402
Haiku 4.5 $0.00201 $0.00201

Measured 5d ago against content hash 5a6fafff071e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

ruleProb 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.

CLAUDE.md · 164 lines

How it starts

The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

1. Project Overview

RuleProbe is a CLI tool for extracting, testing, and validating repository rules from instruction files (CLAUDE.md, AGENTS.md, .cursor/rules, Copilot, etc). It generates testable rules, creates sandbox scenarios, runs multiple providers (mock, dry-run, openrouter, gemini, claude-code, anthropic, openai, ollama, local), and evaluates agent behavior against repo rules. Current version: v2.12.0. Core is stable; AI extraction and hybrid flows are still brittle.

2. Architecture Summary

  • CLI entry: src/cli/index.ts
  • Discovery: src/config/load.ts
  • Extraction: src/rules/extract.ts (deterministic), src/extractors/aiAssisted.ts (AI/hybrid)
  • Scenario generation: src/scenarios/generate.ts
  • Providers: src/providers/* (mock, dry-run, openrouter, gemini, claude-code, anthropic, openai, ollama/local)
  • Matrix command: src/cli/commands/matrix.ts + src/matrix/build.ts + src/reporters/matrix.ts
  • Rule optimizer: src/optimize/* (detect.ts, rewrite.ts) + src/history/ruleHistory.ts
  • Sandbox capture: src/actions/execute.ts captureMode + VirtualOp type
  • Evaluation: src/evaluator/score.ts
  • Reporting: src/reporters/* (json, markdown, html, sarif, junit, pr-comment)
  • Advisor (v1.8.0+): src/advisor/* (repoScan, historyMiner, suggest)
  • Live monitoring (v1.11.0+): src/live/* (watcher, eventBus)
  • Integrations (v2.0.0+): src/integrations/* (express, next, serve)
  • Benchmarks/tests: benchmarks/, tests/

3. Core Files to Inspect First

  1. src/rules/extract.ts
  2. src/extractors/aiAssisted.ts
  3. src/scenarios/generate.ts
  4. src/evaluator/score.ts
  5. src/providers/normalize.ts
  6. src/types/index.ts
  7. tests/benchmark.test.ts

4. Working Rules

  • Read CLAUDE.md first
  • Do not ask questions
  • Do not redesign the whole project
  • Do not rewrite architecture unnecessarily
  • Preserve existing passing behavior
  • Verify commands before claiming success
  • Each phase = own branch → pnpm typecheck && pnpm test green → merge main → tag vX.Y.Z → npm auto-publish

Read the full file on GitHub · 164 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. 5d ago First seen · 164 lines · 2,010 tokens per session scan A 5a6fafff071e

Subscribe to this mod's changes

ruleProb CLAUDE.md is an instructions file published in the GitHub repository canblmz1/ruleProb (7 stars, last pushed 3mo ago), licensed MIT. It adds 2,010 tokens to every session, about $0.0100 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.