A specialized utility designed to identify and remove AI-generated code slop, redundant defensive logic, and inconsistent styling from git branches.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deslop
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install deslop using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Deslop?
The Deslop skill is an essential tool for developers using AI-assisted coding workflows who want to maintain a clean, human-readable codebase. AI agents often introduce 'slop'—excessive defensive checks, unnecessary type casts, redundant comments, and lint ignores—that can clutter a repository. This skill specifically targets these artifacts by analyzing git diffs and applying project-specific heuristics to prune away noise without affecting core functionality.
By leveraging Deslop within your Openclaw Skills library, you ensure that AI-generated contributions remain high-quality and consistent with your team's local style conventions. It acts as a final polish layer, bridging the gap between raw AI output and production-ready code by preserving domain-valid guards while stripping away the robotic residue often left behind by Large Language Models.
Deslop Use Cases
- Post-AI code generation cleanup to remove unnecessary defensive noise and verbose comments.
- Standardizing feature branches before submitting pull requests for human review.
- Refining code style by eliminating redundant casts and placeholder debug leftovers.
- Auditing branch diffs for inconsistent lint-ignore directives and placeholder logic.
How Deslop Works
- Comparison: The skill sets a comparison base (defaulting to main) and inspects the git diff against the current HEAD.
- Candidate Identification: It utilizes ripgrep (rg) to scan added lines for common slop patterns such as redundant catches, placeholder comments, and debug remnants.
- Contextual Review: Every candidate is reviewed within its full file context to determine if it aligns with or deviates from local project patterns.
- Pruning: The agent removes only the inconsistent slop while strictly preserving security boundaries and valid domain guards.
- Verification: Project-specific checks like bun check or bun typecheck are executed to ensure no regressions were introduced.
- Reporting: A final summary is provided, detailing exactly which files were modified and what logic was intentionally kept for safety.
Deslop Setup
To integrate Deslop into your local environment for Openclaw Skills, ensure you have git and ripgrep installed. You must also include a heuristics reference file in your project.
# Ensure ripgrep is installed
brew install rg
# Ensure you have a heuristics file at
# ./references/slop-heuristics.md
# Fetch the latest base branch for accurate diffing
git fetch origin main
Deslop Data Schema & Taxonomy
The skill manages its operation through git metadata and local configuration files. Its behavior is governed by the following schema:
| Component | Type | Description |
|---|---|---|
| Comparison Base | Git Branch | The reference point for identifying new code (default: main) |
| Slop Heuristics | Markdown | A reference file located at references/slop-heuristics.md containing pattern rules |
| Tool Dependencies | Binary | Requires Bash and ripgrep (rg) for text processing |
| Integrity Checks | Command | Uses project-level scripts like bun check to validate changes |
Deslop Advanced Features
- Intelligent Guardrails: Automatically detects and preserves protections at trust boundaries like authentication, database, and network I/O.
- Minimal Edit Preference: Prioritizes surgical removals over broad code rewrites to minimize merge conflicts.
- Integrated Regression Testing: Hooks directly into your local type-checkers and linters to verify code health after cleanup.
- Context-Aware Filtering: Differentiates between generic AI slop and necessary project conventions when using Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Cron Job Setup
A professional framework for creating and managing reliable, scheduled AI tasks and periodic workflows within OpenClaw.
Git Commit & Push Automator
Automatically stage changes, generate AI-driven commit messages, and push code to your remote repository in one seamless workflow.
Caddy Wildcard Reverse Proxy
A specialized tool for managing Caddy-based reverse proxy routes with wildcard subdomain support and automated HTTPS for local services.
Amazon Automation
A browser automation skill for purchasing, reordering, and processing returns on Amazon using an AI agent.
dev-serve
A command-line utility to instantly expose local development projects via Caddy-managed subdomains with automated tmux session management.
Domain Check & Buy (Vercel)
A comprehensive toolkit for checking domain availability, pricing, and managing purchases directly via Vercel and Openclaw Skills.