Back to skill
Skillv3.0.0
ClawScan security
Psyvector Pv34 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 18, 2026, 7:15 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- This instruction-only skill is internally consistent with its stated purpose (a rapid-mediation persona) and requests no credentials, binaries, or installs, so it presents a low security surface.
- Guidance
- This skill is instruction-only and asks for no credentials, so its immediate risk is low. Still: 1) Confirm 'clawhub install' is the official installer on your platform before running it; 2) Review any runtime prompts the installer may show (pricing/payment, license, or network access); 3) Test the skill in a non-production environment first to verify behavior; 4) If you require strict data handling, ask the provider what data (if any) the skill sends off-agent and whether it stores conversation history.
Review Dimensions
- Purpose & Capability
- okName/description (rapid mediation) match the SKILL.md content (persona, response parameters). There are no requested env vars, binaries, or unrelated capabilities.
- Instruction Scope
- okThe SKILL.md contains only persona description, configuration parameters, and a single install hint for 'clawhub install'; it does not instruct reading files, accessing credentials, or contacting external endpoints.
- Install Mechanism
- okNo install spec is present in the registry entry and no code is included. The SKILL.md's example 'clawhub install' is a platform command and not an embedded download; no archives, URLs, or package installs are present.
- Credentials
- okThe skill declares no environment variables, credentials, or config paths. There is no disproportionate request for secret access.
- Persistence & Privilege
- okThe skill is not forced-always, and default autonomous invocation is allowed (platform default). It does not request elevated or persistent system-wide privileges.