clean-code
Remove AI-generated code slop from the current diff while preserving behavior.
Skill by paulorsv01
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/paulorsv01-clean-code
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. Checked Oct 25, 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/clean-code
What's inside
- Skill name
clean-code- Description
- Remove AI-generated code slop from the current diff while preserving behavior. Use for clean-code, code cleanup, unnecessary defensive code, weak casts, excess comments, or avoidable nesting.
- License
- MIT
- Instructions
- 30 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- openai.yaml · agents/
SKILL.md (first lines)
# Clean Code Clean the changed code in place. Preserve behavior and repository conventions. A Clean Code request authorizes focused edits and their verification, not unrelated refactors, commits, pushes, or external writes. ## Scope 1. Resolve the diff from the user's paths, branch, commit, or current changes against the correct base. 2. Read repository instructions and enough surrounding code to establish the local style. 3. Limit edits to slop introduced or exposed by the change. Leave unrelated existing code intact. ## Remove - Comments that restate code, narrate obvious steps, or conflict with local style. - Defensive checks and `try`/`catch` blocks that add impossible states to trusted paths. - `any` casts and assertions used only to bypass a type problem. - Deep nesting that direct control flow or early returns can remove. - Redundant wrappers, helpers, branches, temporary variables, and duplicated logic. - Names, prose, and abstractions that do not match the surrounding codebase. ## Guardrails - Keep behavior unchanged unless the diff contains a clear bug within the requested scope. - Prefer deletion and focused edits over broad rewrites. - Preserve intentional validation at trust boundaries. - Run the smallest authoritative formatter, typecheck, test, or build checks that cover the edits. - Inspect the final diff for behavior drift and unrelated churn. Return what changed, what verification passed, and any residual uncertainty in three concise sentences or fewer.
From paulorsv01/agents, MIT.
More from paulorsv01/agents
- 0.529score
figma-read
paulorsv01
Read a Figma design offline from a .fig export — layer tree, copy, component properties, layout, colors, design-token names, and original vector or raster…
SkillBFreshness: BClaude Code +9 - 0.526score
maintainability-review
paulorsv01
Run an unusually strict maintainability review for structural simplification, abstraction quality, file growth, branching, types, boundaries, and codebase…
SkillBFreshness: BClaude Code +9 - 0.526score
deep-review
paulorsv01
Run independent correctness and maintainability audits, then synthesize one evidence-backed verdict.
SkillBFreshness: BClaude Code +9 - 0.524score
domain-modeling
paulorsv01
Define domain terminology or record architectural decisions when the user requests domain modeling or decision documentation.
SkillBFreshness: BClaude Code +9 - 0.523score
diagnosing-bugs
paulorsv01
Investigate bugs with uncertain causes, intermittent failures, or performance regressions using reproducible evidence.
SkillBFreshness: BClaude Code +9 - 0.523score
gh-stack
github
Create, inspect, or update stacks of dependent branches and PRs with the gh-stack CLI extension.
SkillBFreshness: BClaude Code +9 - 0.521score
cli-for-agents
paulorsv01
Design or review command-line interfaces for reliable agent use.
SkillBFreshness: BClaude Code +9 - 0.521score
correctness-review
paulorsv01
Audit a branch or pull-request diff for correctness, security, breaking behavior, developer-experience regressions, and feature-gate leaks.
SkillBFreshness: BClaude Code +9 - 0.518score
codebase-design
paulorsv01
Design module interfaces or evaluate module depth when reviewing architecture, coupling, or testability.
SkillBFreshness: BClaude Code +9 - 0.518score
research
paulorsv01
Investigate a technical question using primary sources and save a cited Markdown research note.
SkillBFreshness: BClaude Code +9 - 0.518score
claude-advisor
paulorsv01
Consulte o Claude Code para uma segunda opinião somente quando o usuário pedir explicitamente.
SkillBFreshness: BClaude Code +9 - 0.518score
grill-me
paulorsv01
Start the grilling interview when the user invokes grill-me to sharpen a plan or design.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 24, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/paulorsv01/agents
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from paulorsv01
- 0.529score
figma-read
paulorsv01
Read a Figma design offline from a .fig export — layer tree, copy, component properties, layout, colors, design-token names, and original vector or raster…
SkillBFreshness: BClaude Code +9 - 0.526score
maintainability-review
paulorsv01
Run an unusually strict maintainability review for structural simplification, abstraction quality, file growth, branching, types, boundaries, and codebase…
SkillBFreshness: BClaude Code +9 - 0.526score
deep-review
paulorsv01
Run independent correctness and maintainability audits, then synthesize one evidence-backed verdict.
SkillBFreshness: BClaude Code +9 - 0.524score
domain-modeling
paulorsv01
Define domain terminology or record architectural decisions when the user requests domain modeling or decision documentation.
SkillBFreshness: BClaude Code +9 - 0.523score
diagnosing-bugs
paulorsv01
Investigate bugs with uncertain causes, intermittent failures, or performance regressions using reproducible evidence.
SkillBFreshness: BClaude Code +9 - 0.521score
cli-for-agents
paulorsv01
Design or review command-line interfaces for reliable agent use.
SkillBFreshness: BClaude Code +9
Cite this page
clean-code skill. Agent Mart. https://agentmart.dev/skills/paulorsv01-clean-code (last updated 2026-10-24)