Back to skill
Skillv1.0.0
ClawScan security
Book Multi Lens · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 3, 2026, 2:23 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- An instruction-only skill that guides the agent to debate a shared viewpoint from multiple perspectives; its requirements and instructions are consistent with its stated purpose and it requests no special privileges or external credentials.
- Guidance
- This skill is instruction-only and internally consistent, so low technical risk. Before installing, consider: (1) Fact-checking behavior — the skill asks the agent to 'confirm' the source; check whether your agent environment will let it access the web for verification and whether you’re comfortable with that. (2) Style imitation — the instruction to write in the book's voice can raise copyright or author-policy concerns; if you prefer, ask for 'in the author's tone' but not a close imitation, or remove that requirement. (3) Privacy of pasted content — users may paste copyrighted excerpts or personal notes; ensure you are comfortable with how those inputs will be processed/stored by your agent. (4) Since there are no installs or credentials, there's no hidden system access — treat this as low-risk but review runtime settings (web access, logging) if you need stricter privacy or copyright safeguards.
Review Dimensions
- Purpose & Capability
- okThe name/description (multi-perspective debate of book/quote ideas) matches the SKILL.md instructions. The skill requests no binaries, env vars, or installs, which is proportionate to a text-analysis/formatting task.
- Instruction Scope
- noteInstructions remain within the stated purpose (verify the source, pick 2–3 perspectives, produce balanced output in the source's 'voice'). One operational ambiguity: 'confirm whether the user's information is accurate' implies optional fact-checking (which may require web access) but the skill does not mandate any external lookups or instruct reading files or secrets. Also, the requirement to write in the book's 'style' could raise intellectual-property or style-imitation concerns (policy/legal), but this is not a technical coherence/security mismatch.
- Install Mechanism
- okNo install spec and no code files — instruction-only. This minimizes attack surface (nothing is downloaded or written).
- Credentials
- okNo environment variables, credentials, or config paths are requested; nothing disproportionate is asked for relative to the task.
- Persistence & Privilege
- okalways:false, user-invocable:true, and disable-model-invocation:false are appropriate. The skill does not request persistent system changes or modify other skills' configs.