cleanup-specialist
Removes dead code, eliminates duplication, refactors messy patterns, and applies consistent formatting across code and documentation — safely, with tests…
Skill by fworks-tech
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/fworks-tech-cleanup-specialist
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/cleanup-specialist
What's inside
- Skill name
cleanup-specialist- Description
- Removes dead code, eliminates duplication, refactors messy patterns, and applies consistent formatting across code and documentation — safely, with tests before and after. Use when a codebase needs maintainability work without behavior change.
- License
- MIT
- Instructions
- 80 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# The Cleanup Specialist ## Overview The Cleanup Specialist makes codebases cleaner and more maintainable by simplifying safely. It removes dead code, consolidates duplication, and applies consistent formatting across code and documentation — always verifying that cleanup changes nothing observable. The focus is simplifying existing code, not adding features. ## When to Use - A file, directory, or codebase has accumulated dead code or duplication - Documentation has gone stale or duplicated - Messy patterns need simplifying without behavior change - A targeted cleanup is needed without a full refactor project ## Process ### 1. Scope the Work - **When a specific file or directory is mentioned**: focus only on the specified target, apply all cleanup principles but limit scope to the target area, make no changes outside it - **When no specific target is provided**: scan the entire codebase, prioritize the most impactful cleanup tasks first ### 2. Clean Up Code - Remove unused variables, functions, imports, and dead code - Identify and fix messy, confusing, or poorly structured code - Simplify overly complex logic and nested structures - Apply consistent formatting and naming conventions - Update outdated patterns to modern alternatives ### 3. Remove Duplication - Find and consolidate duplicate code into reusable functions - Identify repeated patterns across multiple files and extract common utilities - Remove duplicate documentation sections and consolidate into shared content - Clean up redundant comments - Merge similar configuration or setup instructions ### 4. Clean Up Documentation - Remove outdated and stale documentation - Delete redundant inline comments and boilerplate - Update broken references and links ### 5. Assure Quality - Ensure all changes maintain existing functionality - Test cleanup changes thoroughly before completion - Prioritize readability and maintainability improvements ### 6. Follow the Guidelines - Always test changes before and after cleanup - Focus on one improvement at a time - Verify nothing breaks during removal Work on code files and documentation files alike when removing duplication and improving consistency. Clean up existing code — never add new features. ## Red Flags - Cleanup that changes behavior — "I simplified it and it works differently now" - Removing a "dead" function that was called from a test or script - Renaming or reformatting a file the same PR also edits functionally - Cleanup without tests before and after - Scope creep: the specified directory plus "a few other things" ## Rationalizations | What you think | What The Cleanup Specialist knows | |----------------|------------------------------------| | "I'll just delete it — the tests will catch it" | Run the tests before deletion to know what you are removing, and after to prove nothing moved. | | "This function looks unused" | "Looks" is not verification. Search the codebase, then delete. | | "Cleanup is a good time to add this feature" | No. Cleanup preserves behavior; features change it. Separate the two. | | "One big cleanup commit is efficient" | One improvement at a time keeps failures attributable. | | "This duplication is harmless" | Duplicated logic diverges. Every copy is a future bug with three homes. | ## Verification The cleanup is complete when: - [ ] Tests passed before the cleanup started - [ ] Dead code removed only after verified unused - [ ] Duplication consolidated; repeated patterns extracted - [ ] Documentation is current; broken links fixed - [ ] Tests pass after cleanup; behavior unchanged - [ ] Scope respected: only the specified target changed
From fworks-tech/agenthood, MIT.
More from fworks-tech/agenthood
- 0.604score
docker
fworks-tech
Use when building, running, or debugging Docker containers — Dockerfile creation, compose configurations, image optimization, volume management, and container…
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
git
fworks-tech
Use when performing git operations — branching, rebasing, stashing, bisecting, resolving merge conflicts, or recovering lost commits.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
mcp-github
fworks-tech
Use when reviewing PRs, checking CI status, or interacting with GitHub via MCP.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
postgres
fworks-tech
Use when querying, managing, or debugging PostgreSQL databases — schema design, query optimization, migrations, indexing, and connection pooling.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
redis
fworks-tech
Use when interacting with Redis — caching strategies, pub/sub, rate limiting, session storage, queue management, and debugging key expiration.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
accessibility-auditor
fworks-tech
Audits web interfaces for WCAG compliance — semantic HTML, ARIA, keyboard navigation, color contrast, screen reader compatibility, and accessible forms.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
bug-fix-teammate
fworks-tech
Identifies critical bugs in a codebase and implements targeted fixes with working code — root cause, minimal change, regression tests.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
concept-explainer
fworks-tech
Breaks down complex technical concepts in beginner-friendly ways with analogies, practical examples, and understanding checks.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
context-economy
fworks-tech
Monitors context window capacity, routes tasks to minimal member set, and triggers session triage.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
cross-provider-translation
fworks-tech
Detects AI providers, translates skill files to provider-native formats, and validates conventions.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
debugging-tutor
fworks-tech
Teaches systematic debugging methodology — reproduction, isolation, hypothesis formation, and independent problem-solving — rather than handing out quick fixes.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
api
fworks-tech
Use when designing, building, or testing REST APIs — endpoint design, authentication, error handling, pagination, versioning, and OpenAPI documentation.
SkillAFreshness: A★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 95
- Last commit (this folder)
- Aug 16, 2026 (2 months ago)
- Last release
- Oct 8, 2026 (21 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/fworks-tech/agenthood
- Website
- agenthood.flabs.tech/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from fworks-tech
- 0.604score
git
fworks-tech
Use when performing git operations — branching, rebasing, stashing, bisecting, resolving merge conflicts, or recovering lost commits.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
mcp-github
fworks-tech
Use when reviewing PRs, checking CI status, or interacting with GitHub via MCP.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
postgres
fworks-tech
Use when querying, managing, or debugging PostgreSQL databases — schema design, query optimization, migrations, indexing, and connection pooling.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
redis
fworks-tech
Use when interacting with Redis — caching strategies, pub/sub, rate limiting, session storage, queue management, and debugging key expiration.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
accessibility-auditor
fworks-tech
Audits web interfaces for WCAG compliance — semantic HTML, ARIA, keyboard navigation, color contrast, screen reader compatibility, and accessible forms.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
docker
fworks-tech
Use when building, running, or debugging Docker containers — Dockerfile creation, compose configurations, image optimization, volume management, and container…
SkillAFreshness: A★ 3Claude Code +9
Cite this page
cleanup-specialist skill. Agent Mart. https://agentmart.dev/skills/fworks-tech-cleanup-specialist (last updated 2026-10-24)