A safe and automated workspace cleaning skill that inventories, reviews, and archives temporary build files, caches, and logs in coding repositories.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-workspace-hygiene
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 agent-workspace-hygiene 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 Workspace Hygiene?
Workspace Hygiene is a specialized tool designed to keep your development environment tidy, efficient, and free of clutter without rewriting or endangering your project code. Operating seamlessly within repository-based workflows, this skill helps developers and AI agents execute safe, non-destructive cleanups of untracked build artifacts, temporary caches, test reports, and log files. By prioritizing safety and keeping operations quiet and cheap, it ensures your project directories remain optimized without interrupting normal coding workflows.
As part of the broader suite of Openclaw Skills, Workspace Hygiene introduces a recovery-first approach. Instead of permanently deleting your data, it default-archives approved items to a local directory with a detailed manifest. This ensures that any accidental cleanups can be immediately restored, giving developers total peace of mind when executing dry runs or automated disk-space cleanups.
Workspace Hygiene Use Cases
- Freeing up disk space by cleaning heavy build outputs like node_modules/.cache, .next/cache, or dist folders.
- Running dry-run inventories before major git commits to identify untracked or ignored junk files.
- Archiving test-related clutter, including playwright-reports, coverage data, and temporary databases.
- Safely managing and review-clearing scratch files generated by autonomous AI agents during development.
- Organizing legacy development repositories that have accumulated stale .bak, .tmp, or .rej files over time.
How Workspace Hygiene Works
- Trigger on Demand: The skill remains completely silent during ordinary development and only activates when you explicitly request a workspace cleanup, dry run, or disk inventory.
- Locate Workspace Root: It accurately identifies the current repository root to avoid scanning or affecting your home directory or unrelated neighbor projects.
- Run a Cheap Inventory: Using an optimized Python helper script, it scans the repository to categorize artifacts into safe-to-clean, review-required, or protected statuses without heavy computation.
- Generate Compact Plan: The system presents a minimal summary outlining the findings, specifying what is safe to remove, what needs approval, and what is protected.
- Recoverable Archive: Upon your explicit confirmation, the skill packages safe candidates into a timestamped
.codex-cleanup/archive/directory and writes an audit manifest. - Post-Cleanup Verification: It double-checks git status or directory trees to verify space savings and confirms the successful execution of the cleanup plan.
Workspace Hygiene Setup
To configure and run the Workspace Hygiene skill, ensure you have Python and Git available in your environment. You can initialize the inventory check or run a dry-run cleanup directly via the command line.
Execute the workspace inventory script to inspect target artifacts safely:
python ./scripts/workspace_inventory.py --root . --summary-json --max-entries 40 --recent-hours 24
To apply automated project-specific skip rules, you can define an optional configuration file named .workspace-hygiene.yml in your repository root with customized keep/skip patterns.
Workspace Hygiene Data Schema & Taxonomy
The Workspace Hygiene skill organizes all generated logs, cleanup actions, and archived files inside a standardized local directory structure to ensure transparency and easy recovery.
Directory Structure
| Directory/File | Description |
|---|---|
.codex-cleanup/ |
Main directory for all cleanup-related outputs. |
.codex-cleanup/archive/<timestamp>/ |
Contains the raw archived directories and files safely moved out of active workspace roots. |
.codex-cleanup/archive/<timestamp>/manifest.json |
An audit log cataloging files before removal. |
Manifest Taxonomy
The manifest.json preserves critical metadata for every moved artifact:
- original_path: The source file location before archival.
- size: The file size in bytes to calculate exact disk savings.
- modified_time: Last modification date to help with manual restoration.
- action: The specific operation executed (e.g.,
archived,skipped).
Workspace Hygiene Advanced Features
- Smart Manifest Archival: Moves target files to a timestamped archive path rather than executing destructive hard-deletions, ensuring cleanups are fully reversible.
- Configuration Overrides: Supports a
.workspace-hygiene.ymlconfig file to define custom ignore, keep, or automated skip filters for complex repository layouts. - Secret & Credential Isolation: Automatically protects files containing potential credentials, tokens, and local configuration environments like
.envwithout exposing path names. - Integration with Openclaw Skills: Designed to coordinate with other agentic tools to keep logs clear and workspace footprints light during continuous integration runs.
SKILL.md
Loading
Related Openclaw Skills
Learning Path Designer
A structured corporate curriculum planning assistant that converts employee competency gaps into fully mapped learning paths with Bloom's taxonomy objectives.
Long Task Handoff
An ultra-light automatic handoff manager that preserves critical session state across context-boundary moments and agent restarts.
Workers' Comp Claim Intake AI Assistant
An AI-powered intake tool that guides HR managers and safety officers through documenting workplace injuries, determining OSHA recordability, and drafting state-specific First Report of Injury (FROI) forms.
Pentest Findings Report
An automated reporting assistant that structures raw security findings into industry-compliant penetration test reports.
Geriatric Care Needs Assessment
An interactive, standard-aligned assessment tool designed to help eldercare managers and social workers draft comprehensive care needs reports step by step.
Model Card Drafter
An interactive AI-guided workflow for drafting compliant, structured Model Cards for machine learning models.
Featured*
