Back to skill
Skillv1.9.72
ClawScan security
Wip Readme Format · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 21, 2026, 9:25 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requirements, install mechanism, and runtime instructions match its stated purpose (reformatting a repository README) and do not request unrelated credentials or system access.
- Guidance
- This tool appears to do what it says: it reads repo files (README, SKILL.md, package.json, LICENSE, etc.), generates staged section files, and can assemble them into README.md and TECHNICAL.md. Before installing or running: 1) Only run it on repositories you trust or where you want automated edits, because it will write files (unless --dry-run is used). 2) Note the SKILL.md suggests running other commands like wip-install; review any secondary tools before allowing them to run. 3) The package is installed from npm—confirm you trust the @wipcomputer package source and version. If you want extra safety, run with --dry-run or in a fork/branch so you can review changes before committing.
Review Dimensions
- Purpose & Capability
- okName/description (README formatter) align with the actual code and SKILL.md: it requires Node, reads package.json and SKILL.md, generates section files, and assembles or validates README.md. The single declared dependency (node) and the npm package install are appropriate for this purpose.
- Instruction Scope
- okRuntime instructions and the bundled format.mjs operate only on repository files (README, SKILL.md, package.json, LICENSE, TECHNICAL.md, optional .license-guard.json, etc.) and generate README-init-*.md and TECHNICAL.md. The SKILL.md and code do reference external links (GitHub, shields.io) for badges and documentation, but do not instruct data exfiltration or reading unrelated system paths or environment variables.
- Install Mechanism
- okInstall is via npm package @wipcomputer/wip-readme-format which is a standard registry mechanism. No downloads from arbitrary URLs or extract-from-unknown-host operations are present in the provided metadata.
- Credentials
- okThe skill declares no required environment variables, no credentials, and no config paths. The code only reads/writes files inside the target repository; requested access is proportional to the stated task.
- Persistence & Privilege
- okalways:false (not forced into all agents). It allows autonomous invocation (platform default), but that is not combined with broad credential access or system-wide configuration changes. The tool writes files inside target repos (expected behavior) and does not modify other skills or global agent settings.