Back to skill
Skillv0.1.0
ClawScan security
Skill Reviewer · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 12, 2026, 12:56 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- This is an instruction-only skill that provides a rubric and procedures for auditing other skills; it asks for no installs, secrets, or external endpoints and its requirements align with its stated purpose.
- Guidance
- This skill is instruction-only and internally consistent: it provides a rubric and reads local skill files and bundled references to perform reviews. Before using it, ensure (1) you only have the agent inspect skill folders you are comfortable sharing (SKILL.md or references sometimes contain sensitive notes), (2) your agent's file-access permissions are limited to intended locations, and (3) if you enable strict mode the reviewer will read the bundled full guide (references/anthropic-skills-development-guide.md). Because there is no code to execute or external installs, the primary risk is accidental disclosure of any sensitive content present in the skill files you ask it to review.
Review Dimensions
- Purpose & Capability
- okThe skill's name/description (Skill Reviewer) match its contents: a checklist, scoring rubric, and references for auditing other skills. It does not request unrelated binaries, environment variables, or permissions that would be out of scope for a reviewer.
- Instruction Scope
- okSKILL.md instructs the agent to read SKILL.md and local references and to run simple file-inspection examples (ls, head) against skill folders — all coherent for a reviewer. It does not direct the agent to exfiltrate data or access unrelated system credentials. Note: it assumes the agent can read the target skill files (expected for an auditor).
- Install Mechanism
- okNo install spec and no code files are included (instruction-only), so nothing is written to disk or downloaded during install.
- Credentials
- okThe skill declares no required environment variables, credentials, or config paths. The references and examples are documentation and checklist items only — no secrets requested.
- Persistence & Privilege
- okalways is false and the skill does not request persistent system changes or to modify other skills' configs. It can be invoked by the agent as normal; nothing in the instructions grants elevated permanent privileges.