Voom
The recursive agent containment and cleanup skill. Use this skill whenever an AI agent, pipeline, or autonomous workflow has gone off the rails — producing cascading side effects, unintended outputs, runaway processes, or metaphorical "pink stuff" spread across systems. Triggers include: agent loops, hallucination cascades, unintended bulk writes, recursive failures, agentic sprawl, pipeline contamination, "it's doing something weird", "everything is broken", "make it stop", or any situation where one agent's output has contaminated another agent's input. Also use when the user says /voom. Thi
Install
npx skills add roby2358/skills --skill voomgit clone --depth 1 https://github.com/roby2358/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/voom ~/.claude/skills/voomPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# The /voom Skill ## Background In Dr. Seuss's "The Cat in the Hat Comes Back," a pink spot spreads from a bathtub to a dress to a wall to shoes to a bed — each cleanup attempt by a smaller cat (A through Z) merely transfers the problem elsewhere, until Little Cat Z unleashes VOOM: a mysterious substance that cleans everything at once. This is the canonical agent containment narrative. Your agents have spread pink stuff everywhere. This skill is your VOOM. ## When to Use You are in a /voom situation if: - An agent wrote outputs that became inputs to other agents, spreading errors - A pipeline is producing cascading failures across multiple systems - Bulk operations went wrong and partial results are scattered everywhere - An LLM agent hallucinated confidently and downstream systems acted on it - You can see the pink stuff but every time you clean one spot, another appears - The user says "make it stop" with a haunted look in their eyes ## The Protocol ### Phase 1: CONTAINMENT (Little Cat A) Stop the bleeding. Before cleaning anything, prevent further spread. ``` 1. Identify all running agents/processes involved 2. Pause, kill, or disable them — newest first, outermost first 3. Document what's still running and what's been stopped 4. Identify the blast radius: what systems/files/databases were touched? ``` **Output a Contamination Map**: List every system, file, database, API, or service that the pink stuff has reached. Be explicit. Miss nothing.
Details
- Skill name
voom- Repository
- roby2358/skills
- Path
voom/SKILL.md- Repo updated
- 14 November 2026
FAQ
What does the Voom skill do?
The recursive agent containment and cleanup skill. Use this skill whenever an AI agent, pipeline, or autonomous workflow has gone off the rails — producing cascading side effects, unintended outputs, runaway processes, or metaphorical "pink stuff" spread across systems. Triggers include: agent loops, hallucination cascades, unintended bulk writes, recursive failures, agentic sprawl, pipeline contamination, "it's doing something weird", "everything is broken", "make it stop", or any situation where one agent's output has contaminated another agent's input. Also use when the user says /voom. Thi
How do I install the Voom skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add roby2358/skills --skill voom. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Voom popular?
It lives in roby2358/skills, which has 5 GitHub stars and 0 forks.
Related skills
Coding
5roby2358/skills
Project-specific coding conventions and constraints. Use when writing, modifying, or refactoring any code, implementing features, writing tests, or building MCP servers in this project.
Game Design
5roby2358/skills
Design principles for tactics and strategy games. Use when designing new mechanics, tuning existing systems, adding units or abilities, adjusting difficulty, or making any game design decision. Triggers on discussions about game balance, unit roles, terrain, pacing, spawning, healing, combat, win/loss conditions, or UI for conveying game state.
Game Productize
5roby2358/skills
Audit a browser game for ship-readiness and produce a tiered gap list: a SHIP.md in the game repo plus pjpd tasks. Finds what gets forgotten between "the game works" and "a stranger can play it": title/menus/end screens, save and resume, audio and settings, onboarding, the mobile "sorry" screen, localization readiness, error recovery, hosting on the author's site and on portals (itch.io, Newgrounds), credits, asset licenses, and AI disclosure. Use when the user says "productize", "/game-productize", "get this game ready to ship/release/publish/launch", "what's missing before I launch", "prep f
Webmcpify
9sandbaseai/workbuddy-skill
Make a web app agent-ready — propose a WebMCP tool manifest, integrate, verify in a real browser, heal; unrelated code stays untouched. Use for "webmcpify", "add WebMCP", or "expose app actions to AI agents".
Verify Agent Action
9sandbaseai/workbuddy-skill
Use before a consequential agent action to verify its exact target, parameters, scope, approval binding, evidence, freshness, and replay safety without executing it.
Conway Method
5tilmon-engineering/claude-skills
Use when working in a project that has a `.agents/` directory or when the user asks about Conway Architecture, the conway-architecture plugin, domain agents, owned_paths enforcement, or how to coordinate a persistent subagent team that shadows the application's architecture
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.