persistproc: Instructions file for Claude Code

CLAUDE.md

persistproc CLAUDE.md is an instructions file for Claude Code from irskep/persistproc. It costs 502 tokens per session, scanned A, original, MIT.

Development instructions for the persistproc Python project, covering its structure, tests, formatting, type checks, and safe ways to investigate problems.

In plain words
What is it for?
Use them when modifying persistproc, running its pytest suite, checking code with Ruff, or debugging through the project’s test helpers and logs.
Why use it?
They reduce the risk of disrupting test servers or misdiagnosing failures during development.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md.

This is irskep/persistproc's own configuration. It tells Claude Code how to work on persistproc 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 persistproc configures →

Reuse

Borrowing it

Nothing to install: this file belongs to irskep/persistproc. 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/irskep/persistproc/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/irskep/persistproc

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 persistproc CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/irskep/persistproc/claude-md/github.svg)](https://agentmods.dev/instructions/irskep/persistproc/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/irskep/persistproc/claude-md"><img src="https://agentmods.dev/badge/instructions/irskep/persistproc/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for persistproc CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/irskep/persistproc/claude-md"><img src="https://agentmods.dev/badge/instructions/irskep/persistproc/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 502 This file is loaded in full into every session.
When invoked 502 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 $0.00502 $0.00502
Opus 5.5 $0.00201 $0.00201
Sonnet 5.5 $0.00100 $0.00100
Haiku 4.5 $0.00050 $0.00050

Measured 2mo ago against content hash 953ebb6370dc, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

persistproc 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 2mo 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 · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Development Notes

Structure

Typically begin by reading README.md.

Key files:

  • persistproc/cli.py
  • persistproc/tools.py
  • persistproc/process_manager.py

Running Tests

Use uv run python -m pytest to run tests. For development, use uv run python -m pytest -x --maxfail=3 to stop after 3 failures.

There is critical code in tests/conftest.py and tests/helpers.py. Some defaults are changed via env vars.

Linting and Type Checking

  • Linting: uv run ruff check
  • Formatting: uv run ruff format

Development Guidelines

NEVER run persistproc commands manually during development. The persistproc CLI should only be invoked through the test suite. Manual CLI usage can interfere with test servers and cause unexpected failures.

Instead:

  • Use the test suite to verify functionality
  • Use the test helpers in tests/helpers.py for programmatic testing
  • Debug issues through test output and logging

NEVER background a process with an & suffix.

ALWAYS use git --no-pager diff for all diffs, never git diff.

If you get a timeout in tests while running locally, it is NOT a "timing issue" or "race condition." It is as REAL BUG. You are running on a fast computer that is not under load. The correct response to a test timeout is to ADD DEBUG LOGGING and THINK HARD.

Workflow guidelines

For EVERY programming task assigned, you are NOT FINISHED until you can produce a message in the following format:

Relevant files found:

  • (list them)

(1-3 paragraphs justifying why the change is both correct and comprehensive)

Steps taken to verify:

  • (list them)

Web links supporting my changes:

  • (list them)

I solemnly swear there are no further steps I can take to verify the changes within the boundaries set for me. I ran the entire test suite just now, which you can see in the output above. If you can't see a complete passing run of all tests, I admit this whole report is bullshit because I didn't do everything you asked.

Read the full file on GitHub · 62 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. 2mo ago First seen · 62 lines · 502 tokens per session scan A 953ebb6370dc

Subscribe to this mod's changes

persistproc CLAUDE.md is an instructions file published in the GitHub repository irskep/persistproc (8 stars, last pushed 1y ago), licensed MIT. It adds 502 tokens to every session, about $0.0020 per session on Opus 5.5. 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.

Related

persistproc CLAUDE.md compared with instructions from other repositories

persistproc CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
persistproc CLAUDE.mdirskep/persistprocInstructions file81y agoMITA502Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode