extract-and-name
What it does
Use when a function exceeds 40 lines or a file exceeds 300. Extracts cohesive sub-units with names that describe intent.
Install opens this entry in your AgentsRoom desktop app. If the app is not installed yet, you will be sent to the download page.
SKILL.md
--- name: extract-and-name description: Use when a function exceeds 40 lines or a file exceeds 300. Extracts cohesive sub-units with names that describe intent. --- # Extract and name When you spot an oversized function or file: 1. Identify cohesive blocks (paragraphs of code that do one thing). 2. Extract each into its own function. Name it after WHAT it does, not HOW. 3. Keep the public surface stable — extraction is internal only. 4. After extraction, the original function should read like a high-level outline. 5. Don't extract for the sake of it: a 60-line linear setup is fine. The signal is "I had to scroll to understand".
Tags
Further reading
Claude Ads: the Claude Code skill that audits your ad accounts
Claude Ads is an open source skill for Claude Code that audits ad accounts on twelve platforms, from Google and Meta to Reddit and X, with evidence for every finding. What v2.1.0 does, how to install it safely, its limits, and how to run it in AgentsRoom.
AGENTS.md: One Context File for Every Coding Agent (Codex, Antigravity, Claude)
AGENTS.md is the portable instructions file your AI coding agents read before touching your code. What to put in it, how it differs from CLAUDE.md, and how to keep one context across Codex, Antigravity and Claude.
Download AgentsRoom
Run all your AI agents, on all your projects, from a single window.
Companion app: monitor your agents on the go
Bring your own: Claude, Codex, Antigravity CLI, or other AI provider.
Push bugs and requests straight to your public backlog.