agentbreak CLAUDE.md

agentbreak CLAUDE.md is an instructions file for coding agents from mnvsk97/agentbreak. It costs 155 tokens per session, scanned A, original, MIT.

A repository instruction document for Claude Code in the AgentBreak project. AgentBreak is a Python package that proxies AI-model and MCP traffic while injecting configurable faults for testing.

In plain words
What is it for?
Use it when working on AgentBreak, especially to initialize or validate its configuration, inspect or serve traffic, test agent resilience, and verify changes.
Why use it?
It gives the coding assistant the project's purpose, setup commands, resilience-testing workflow, and verification requirement in one place.

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/mnvsk97/agentbreak/claude-md
Clone the repo
git clone --depth 1 https://github.com/mnvsk97/agentbreak

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 agentbreak CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/mnvsk97/agentbreak/claude-md.svg)](https://agentmods.dev/instructions/mnvsk97/agentbreak/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/mnvsk97/agentbreak/claude-md"><img src="https://agentmods.dev/badge/instructions/mnvsk97/agentbreak/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 155 This file is loaded in full into every session.
When invoked 155 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 $0.00155 $0.00155
Opus 5 $0.00077 $0.00077
Sonnet 5 $0.00031 $0.00031
Haiku 4.5 $0.00015 $0.00015

Measured 4d ago against content hash fb25c1d58d01, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentbreak 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 4d 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 · 31 lines

What it actually says

Claude Code — AgentBreak

What this repo is

A Python package that proxies LLM (OpenAI + Anthropic) and MCP traffic, injecting configurable faults. Config lives in .agentbreak/, created by agentbreak init.

Quick commands

pip install -e '.[dev]'
agentbreak init
agentbreak validate
agentbreak serve

For MCP: agentbreak inspect then serve.

Skill

Install the portable skill with:

npx skills add mnvsk97/agentbreak --skill agent-resilience-testing --yes

Then ask: Test my agent for resilience.

Verification

Before suggesting a change is done: run agentbreak verify.

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. 4d ago First seen · 31 lines · 155 tokens per session scan A fb25c1d58d01

Subscribe to this mod's changes

agentbreak CLAUDE.md is an instructions file published in the GitHub repository mnvsk97/agentbreak (20 stars, last pushed 3mo ago), licensed MIT. It adds 155 tokens to every session, about $0.0008 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.