OpenClaw Workspace
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill is coherently focused on auditing an OpenClaw workspace, but it intentionally reads and may recommend changes to persistent memory and behavior files.
This skill appears safe to install as an instruction-only workspace auditor. Use it with awareness that it may inspect personal memory and recommend edits to files that shape future agent behavior; ask it to show findings and proposed diffs before changing anything.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may use stored personal preferences, past mistakes, deadlines, people, and project history when making workspace recommendations.
This directs the agent to consult persistent personal memory and daily notes. That is consistent with the workspace-audit purpose, but it involves sensitive context that may be reused across tasks.
Use memory_search on MEMORY.md plus memory/*.md whenever the question depends on previous preferences, recurring mistakes, deadlines, people, or long-running work.
Only use this skill if you are comfortable with it reviewing workspace memory; ask for a scoped audit and inspect any memory-based conclusions before accepting them.
A poor accepted change could make the agent behave differently in later sessions, remember the wrong things, or apply a flawed workflow repeatedly.
The skill targets bootstrap behavior files, memory, and skills, so incorrect recommendations could affect future sessions or repeated workflows. The visible text frames changes as proposals rather than hidden automatic edits.
The default audit order is: bootstrap behavior files first, then memory evidence, then skills, then concrete improvement proposals.
Review proposed diffs before applying them, keep backups of important workspace files, and prefer small reversible changes.