clean-room-loop

clean-room-loop is a skill for Claude Code from whit3rabbit/clean-room-skill. It costs 156 tokens per session (1,612 once invoked), scanned A, original, MIT.

An unattended workflow for reverse-engineering software into a clean implementation, using several in-session coding-agent roles. It is designed for Claude Code and keeps the discussion separate from the background run.

In plain words
What is it for?
Use it to launch clean-room reverse-engineering tasks, pass in the decisions discussed beforehand, and coordinate the workflow's specialist roles.
Why use it?
It lets a coding task run hands-off without using separate pay-per-token command-line API calls. It also keeps source material and implementation work separated to reduce accidental copying.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: reads .claude/ paths; mentions subagents; names the AskUserQuestion tool.

Part of the clean-room plugin — 10 skills, 7 agents shipped together

Good fit Use it to launch clean-room reverse-engineering tasks, pass in the decisions discussed beforehand, and coordinate the workflow's specialist roles.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/whit3rabbit/clean-room-skill/clean-room-loop
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.

Any agent
npx skills add whit3rabbit/clean-room-skill --skill clean-room-loop
Clone the repo
git clone --depth 1 https://github.com/whit3rabbit/clean-room-skill

Made for: Claude Code.

Or install clean-room, the plugin that ships this one along with the rest of its 10 skills, 7 agents.

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 clean-room-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/whit3rabbit/clean-room-skill/clean-room-loop.svg)](https://agentmods.dev/skills/whit3rabbit/clean-room-skill/clean-room-loop)
Your own site
<a href="https://agentmods.dev/skills/whit3rabbit/clean-room-skill/clean-room-loop"><img src="https://agentmods.dev/badge/skills/whit3rabbit/clean-room-skill/clean-room-loop.svg" alt="Measured on agentmods" height="20"></a>
Per session 156 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,612 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium MCP Rug Pull · line 74
    npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
    Fix: Pin the version: npx @scope/[email protected]
How audits are shown
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 $0.00156 $0.01612
Opus 5 $0.00078 $0.00806
Sonnet 5 $0.00031 $0.00322
Haiku 4.5 $0.00016 $0.00161

Measured 8d ago against content hash 813d9e934f0d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

clean-room-loop 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.

skills/clean-room-loop/SKILL.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Clean-room loop (discussion -> clean-room-loop workflow launch)

Conversational front door to the clean-room-loop dynamic workflow (.claude/workflows/clean-room-loop.js). The workflow runs in the background with no way to ask anything mid-run, so the discussion happens HERE; answers pass as args.

Claude Code only. Dynamic workflows are a Claude Code feature. In Pi/Codex/OpenCode the Workflow() call will not exist - see the Fallback step. Only this workflow shortcut is Claude Code specific; the underlying clean-room skills (/clean-room:unattended, clean-room-skill run) work on every supported runtime.

Installed project-local. The workflow script ships to project-local .claude/workflows/ (not global). Workflow({ name }) discovers it from the current project's .claude/workflows/ (or ~/.claude/workflows/ if a personal copy exists). If the current project does not have it, initialize a project-local install before launching - see Step 3.

What this is (and is NOT)

  • It drives the six clean-room roles with the workflow's OWN agent() subagents (in-session, subscription, no claude -p), gating every wall crossing with the real clean-room-skill artifact validate --role leakage + schema hooks.
  • It is a cost-free path with context-level separation, NOT the OS-enforced wall. A workflow cannot set CLEAN_ROOM_* env or install hooks, so nothing stops a clean subagent from reading source off disk except the neutral-artifact discipline + the leakage gate. If the user needs the enforced boundary, use clean-room-skill run --agent-runtime claude instead (that path costs API tokens by design).
  • The workflow READS the authorized source and WRITES clean specs, plans, code, and reports under the external artifact roots. Confirm authorization and paths before launching.

Steps

  1. Get the brief. Take the end goal from the invocation if present. If missing, ask what they are reimplementing and why they are authorized to.

Read the full file on GitHub · 108 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 · 108 lines · 156 tokens per session scan A 813d9e934f0d

Subscribe to this mod's changes

clean-room-loop is a skill published in the GitHub repository whit3rabbit/clean-room-skill (10 stars, last pushed 5d ago), licensed MIT. It adds 156 tokens to every session and 1,612 once invoked, 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens