Back to skill
Skillv1.0.0
ClawScan security
Security Audit by Jason · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 6, 2026, 3:55 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's code, requirements, and runtime instructions are consistent with a local static security-audit tool and do not request unrelated credentials or hidden network access.
- Guidance
- This appears to be a local static-audit tool; it's coherent with its description. Before installing or auto-wiring it into workflows: (1) review audit.py yourself so you trust the logic, (2) run it in an isolated environment (VM/container) the first few times, (3) remember it's heuristic-only and can produce false positives (large data files, minified code), and (4) do not rely on it as a sole defense — use it as a pre-check and perform manual review for critical code. If you plan to auto-run it after every git clone, be aware you must add that hook yourself; the skill does not auto-install that behavior.
Review Dimensions
- Purpose & Capability
- okName/description match implementation: the skill includes a Python script (audit.py) that performs static checks, and the only declared requirement is python3. There are no unrelated credentials, binaries, or config paths requested.
- Instruction Scope
- noteSKILL.md instructs running audit.py against a local repo and suggests inserting it into workflows (after git clone / clawhub install). The script scans files recursively in the provided directory for suspicious patterns and large single-line files; it does not attempt to read environment secrets or reach external endpoints. Note: the README/SKILL.md suggests automatic invocation by OpenClaw but the package contains no install hooks — wiring into CI/clone workflows is manual.
- Install Mechanism
- okNo install spec; this is instruction + script only. The script is local and there are no downloads or extract steps that would pull arbitrary remote code.
- Credentials
- okThe skill requires no environment variables, no credentials, and only needs read access to the scanned directory. No credentials or unrelated external service tokens are requested.
- Persistence & Privilege
- okalways is false and the script does not write persistent configuration or modify other skills. It prints reports to stdout / optional output file only. Enabling automatic invocation would be a user action.