Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Added explicit "fresh eyes" self-review step in the Simplify pass, requiring re-reading all added/modified code for bugs, errors, and brittle logic before edits. - Clarified the skill’s philosophy: emphasizes disciplined, fresh review using peak context before any further coding steps. - No changes to code or functionality; documentation improvement only. - No functional or interface changes. Safe for all users.
What This Skill Does
A post-completion review hook for coding agents that runs three sequential passes on the task diff: simplify (dead code, naming, control flow cleanup), harden (input validation, injection vectors, auth checks, secrets), and document (inline comments on non-obvious decisions). Activates after a non-trivial coding task completes and pauses for human approval before applying structural refactors or security changes.
Runs while the agent still holds full context about the task, catching issues a later reviewer would need to reconstruct from scratch.
When to use it
- Catching unused imports and debug logs after finishing an implementation
- Checking new API endpoints for missing auth guards before opening a PR
- Adding decision comments to non-obvious logic before handing off for review
- Flagging hardcoded secrets or unvalidated user inputs in freshly written code
- Running a bounded quality and security sweep before marking a task done
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Review my changes after completing the user export endpoint
- 1Re-reads all modified files with fresh eyes before making any edits
- 2Applies cosmetic fixes automatically (removes unused imports, tightens naming, flattens control flow)
- 3Presents any structural refactor as an interactive proposal and waits for explicit approval
- 4Checks the new endpoint for input validation gaps, missing role checks, and exposed secrets
- 5Adds up to 5 inline comments on non-obvious decisions and outputs a structured YAML summary
Structured YAML summary listing applied fixes, flagged issues, comments added, budget utilization, and whether follow-up review is required