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/exiw-ai/proofloop/claude-mdgit clone --depth 1 https://github.com/exiw-ai/proofloopWhat 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 | $0.00551 | $0.00551 |
| Opus 5 | $0.00275 | $0.00275 |
| Sonnet 5 | $0.00110 | $0.00110 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
proofloop 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 2d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Instructions for AI coding agents working on this repository.
Quick Reference
make test # Run tests
make pre-commit # Run linters, typechecker, formatters
make check # Run all checks (pre-commit + tests)
Project Overview
Proofloop is an AI agent orchestrator - agents that run until done. Define conditions, go to sleep, wake up to results. Supports multiple providers: Claude Code, OpenCode, and Codex.
Key technologies: Python 3.12+, Typer (CLI), Pydantic, asyncio, Claude Code SDK, OpenCode, Codex
Code Style
- Python 3.12+ with type hints required
- Async/await for I/O operations
- Use
logurufor logging - Follow existing patterns in the codebase
Comments Policy
- Avoid comments that restate the code
- Docstrings only for public interfaces and non-obvious logic
- Goal: self-documenting code
Architecture
The project follows layered architecture:
src/
├── cli/ # Command-line interface
├── domain/ # Business logic, entities, value objects
├── application/ # Use cases, orchestration
└── infrastructure/ # External adapters (Claude SDK, Git, MCP)
Rules:
- Domain layer has no external dependencies
- Infrastructure implements interfaces defined in domain
- Keep business logic in domain, not in adapters
Guidelines
Do
- Follow existing project structure and patterns
- Keep changes minimal and focused
- Add tests for functional changes
- Run
make checkbefore committing
Don't
- Commit secrets or API keys
- Break existing tests
- Add unnecessary defensive code for impossible cases
- Invent custom implementations when a library exists
Dependencies
Core libraries used in the project:
- Validation:
pydantic - Retry logic:
tenacity - Logging:
loguru - HTTP:
httpx
When Done
A change is complete when:
- All checks pass (
make check) - New behavior is covered by tests
- Working tree is clean
System Contract
Key guarantees that must be preserved when modifying the core pipeline:
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.
- 2d ago First seen · 86 lines · 551 tokens per session scan A 60ef0f8bd097
proofloop CLAUDE.md is an instructions file published in the GitHub repository exiw-ai/proofloop (38 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 551 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
Maestro CLAUDE.md
Instructions for aaryansinha16/Maestro, covering maestro — claude.md, project identity, core operating principles, how it works and the conductor (brain).
SpecRoute CLAUDE.md
Instructions for Enovatr-Labs/SpecRoute, a project described as: Open source framework for spec-driven agentic software engineering. PRDs, prompts, agents, skills, workflows, and coding rules for Codex, Claude, and modern AI development teams.
zeroshot CLAUDE.md
Claude Code instructions for the-open-engine/zeroshot, covering zeroshot: multi-agent coordination engine, 🔴 critical rules, 🔴 general purpose requirement (critical), 🔴 behavioral standards and where to look.
map-framework CLAUDE.md
Instructions for azalio/map-framework, covering map framework (mapify-cli) — agent instructions, what this repo is, critical invariant: template single-source render, skill catalog invariant and how to work in this repo.
agentic-os AGENTS.md
Instructions for modimihir07/agentic-os, covering agentic os — complete project context for ai agents, role definition, project identity, architecture and 3-agent engine.
cctop AGENTS.md
Instructions for st0012/cctop, covering agents.md - development guide for cctop, scope and sources, required development rules, driver workflow and github and pr rules.