haircutThe barber for your coding agents' config. Weighs everything every agent on this machine loads (instruction files, memory, MCP servers, hooks, skills, commands, subagents, plugins, leftovers), prices each item in tokens from your own transcripts, and parks the dead weight with a receipt and an undo. Use this skill whenever the user mentions config bloat, a heavy or slow session start, tokens burned before the first word, too many MCP servers, hooks that are slow or fire on every prompt, skills or commands or plugins nobody invokes, cleaning up or auditing CLAUDE.md or AGENTS.md, Codex config, or asks what does my agent actually load, why is my context full before I type, which of these servers can I turn off, or what is this plugin costing me. Use it even when they only say the session feels bloated or startup got slow, and even when they mention /context, /doctor or /skill-doctor.
Install via ClawdBot CLI:
clawdbot install othmanadi/haircutGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://clawhub.ai/user/othmanadiAudited Oct 12, 2026 · audit v1.0
Generated Oct 12, 2026
A solo developer's Claude Code sessions feel sluggish and they suspect every startup is burning tokens on MCP servers they stopped using months ago. They run haircut, get a before-first-word token figure and a ranked list of parked candidates, then confirm parking the dead servers via AskUserQuestion.
An engineering lead notices their team's shared AGENTS.md and CLAUDE.md files have grown bloated with stale conventions, and a slow hook fires on every prompt. haircut scans the repo, prices each instruction block, and produces a report the lead can walk through with the team before parking the dead weight.
A research team running many Codex rollouts discovers that fixed per-rollout tokens and repeated MCP schema costs are inflating spend per model call. They use haircut's Codex scan and optional probe to quantify the schema tax, then trim unused plugins and servers from the shared config.
A developer migrating to a fresh machine realizes their old config carried years of unused skills, commands, subagents, and plugins. They run haircut with --days 30 to see what actually gets used, keep the essentials, and park the rest with a receipt and undo path on the old machine.
A security-minded engineer runs haircut and surfaces ROTATE items — credentials sitting in plain text across config files. They follow the skill's guidance to identify the file without printing the value, rotate keys themselves, and mark items as kept until the secret is removed.
The core haircut script remains free and open-source for individual developers, while a hosted dashboard aggregates scan reports across a team and tracks config weight over time. Teams pay per seat for historical trends and shared decision records.
haircut is licensed or embedded as a built-in config hygiene tool inside a commercial coding-agent IDE or CLI platform. Users get it as part of their existing subscription, and the platform uses it to reduce support load from bloated configs.
An agency or consultancy offers hands-on agent config audits using haircut, delivering a priced report of what each client's machine loads and a remediation plan. The tool accelerates the audit while the humans handle decisions and enterprise policy.
💬 Integration Tip
Run the read-only measure stage first and match the depth of response to the user's actual question — do not walk someone into a cleanup they did not ask for. Write decisions.json with the Write tool (never a heredoc) and default unresolved items to 'later' so nothing moves without an explicit answer.
Scored Oct 12, 2026
Control desktop applications on Windows — launch, close, focus, resize, move windows, simulate keyboard/mouse input, manage processes, control VSCode, read clipboard, and capture screen info. Use when the user wants to interact with any running program, switch windows, type text, press shortcuts, open files in VSCode, manage running processes, or get system display information.
Conduct rigorous, adversarial code reviews with zero tolerance for mediocrity. Use when users ask to "critically review" my code or a PR, "critique my code", "find issues in my code", or "what's wrong with this code". Identifies security holes, lazy patterns, edge case failures, and bad practices across Python, R, JavaScript/TypeScript, SQL, and front-end code. Scrutinizes error handling, type safety, performance, accessibility, and code quality. Provides structured feedback with severity tiers (Blocking, Required, Suggestions) and specific, actionable recommendations.
Pragmatic coding standards for writing clean, maintainable code — naming, functions, structure, anti-patterns, and pre-edit safety checks. Use when writing new code, refactoring existing code, reviewing code quality, or establishing coding standards.
Claude Code integration for OpenClaw. This skill provides interfaces to: - Query Claude Code documentation from https://code.claude.com/docs - Manage subagents and coding tasks - Execute AI-assisted coding workflows - Access best practices and common workflows Use this skill when users want to: - Get help with coding tasks - Query Claude Code documentation - Manage AI-assisted development workflows - Execute complex programming tasks
Plan, draft, version, and refine written content with enforced versioning and quality audits.
Use when writing tests, creating test strategies, or building automation frameworks. Invoke for unit tests, integration tests, E2E, coverage analysis, performance testing, security testing.