File Management by Brian

Organize and maintain AI agent workspaces with structured directories, clear naming, regular audits, dead file detection, and safe cleanup practices.

Security Scan

The skill's files and runtime instructions are coherent with its stated purpose (workspace organization and audits); it requires no credentials, no network access, and its included script only performs local filesystem inspections.

Purpose & Capability

Name and description match the provided materials: guidance docs plus a local audit script that inspects workspace directories, file types, modification times, and sizes. Nothing in the repository requests unrelated capabilities (no cloud credentials, no external services).

Instruction Scope

SKILL.md directs local inspection commands (grep, find, du, crontab checks) and conservative cleanup practices (use trash, commit before deleting). It references OpenClaw-specific paths (~/.openclaw/workspace) and commands (openclaw cron list) which are consistent with an agent workspace management skill. Instructions do not instruct sending data to external endpoints or reading unrelated secret files.

Install Mechanism

No install spec is present (instruction-only plus a small shell script), so nothing is downloaded or written by an installer. The included audit script is small, readable, and performs only local filesystem queries.

Credentials

The skill declares no required environment variables, credentials, or config paths. The actions it performs (examining workspace files and crontab) are proportional to a file-management/audit tool.

Persistence & Privilege

The skill is not always-enabled and does not request to modify other skills or system-wide configuration. It runs locally and contains no autonomous installation hooks.

Assessment

This skill appears coherent and low-risk, but review before running on sensitive or production workspaces: 1) Inspect audit-workspace.sh yourself (it only lists and reports files) and run it with a test/workspace copy first. 2) Ensure you have backups or git savepoints before deleting anything — the documentation already advises this. 3) Note the SKILL.md references OpenClaw-specific paths and the 'openclaw cron list' command; if that tool isn't present, those checks will fail harmlessly. 4) If you plan automated cleanup, require an explicit manual review step rather than automatic deletes. If you want extra assurance, run the audit script with a non-privileged account and confirm no unexpected commands or network connections occur.

File Management Skill

A battle-tested approach to keeping your AI agent workspace organized and maintainable.

Overview

This skill documents the file management system developed through real-world use of OpenClaw. It covers workspace structure, naming conventions, dead file detection, and cleanup practices.

When to Use This Skill

• Onboarding a new agent or setting up a fresh workspace
• Performing periodic workspace audits
• Before making significant changes to workspace structure
• When workspace feels cluttered or disorganized

Core Principles

1. Every File Has a Purpose

• Active files: Scripts, configs, and data in use by cron jobs or agents
• Reference files: Documentation, strategies, and notes
• Archived files: Old versions, completed project artifacts
• Dead files: Abandoned scripts, old experiments, unused utilities

2. Structure Mirrors Function

workspace/
├── memory/          # Daily session logs and working context
├── skills/           # Installed skill directories
├── project-1/        # Project-specific directories
├── project-2/
├── ACTIVE.md         # Currently running projects & priorities
├── DREAMS.md         # Background processing notes
└── ARCHIVED/         # Completed or abandoned projects

3. Naming Conventions

• Scripts: Use .sh for bash, .py for Python, .js for JavaScript
• Logs: End with .log
• Configs: End with .json, .yaml, or .md
• Daily notes: memory/YYYY-MM-DD.md format

4. Audit Regularly

Run workspace audits monthly or after major changes. Use the audit script to identify:

• Dead files (no references from active crons or scripts)
• Large files consuming storage
• Outdated documentation

Quick Audit Commands

# Find files not referenced by any cron or script
grep -r "filename" ~/path/to/workspace/ --include="*.sh" --include="*.py" --include="*.js"

# Find recently modified files
find ~/path/to/workspace -type f -mtime -7

# Check disk usage by directory
du -sh ~/path/to/workspace/*/

Cleanup Best Practices

1. Never delete immediately — use trash instead of rm
2. Document before deleting — note what a file did in memory first
3. Verify before cleanup — confirm no active references
4. Commit before major cleanup — create a revert point

Full Documentation

See FILE-MANAGEMENT.md for the complete reference implementation, including:

• Directory structure explainer
• Active vs archived file definitions
• Dead file detection criteria
• Example cleanup checklists