Skillv1.0.0

ClawScan security

一个会逐渐成长的AI写作者 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 19, 2026, 7:44 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only LLM persona for generating and mildly adapting literary-style writing; it requests no secrets, installs nothing, and its runtime instructions stay within the described purpose.
Guidance: This skill appears internally consistent and low-risk: it only defines a writing persona and adapts based on reader_feedback. Before installing, consider: (1) provenance — the source/homepage is unknown and author is a generic placeholder ('your-name'), so verify trust if you need strict provenance; (2) privacy — any reader_feedback you supply becomes input to generation, so avoid pasting PII or sensitive data; (3) content safety — review sample outputs to ensure tone/style and moderation safety meet your needs (it may produce ambiguous or emotionally subtle content by design); (4) testing — try a few inputs to confirm the evolution behavior works as you expect. If you require stronger guarantees (e.g., offline-only processing, formal author attribution), request those from the publisher or avoid use until provenance is clarified.

Purpose & Capability: okThe name/description (a 'junior' persona that can evolve with feedback) matches the included files: prompt, schema, and SKILL.md describe a writing persona and feedback-driven micro‑adjustments. There are no unrelated environment variables, binaries, or install steps requested.
Instruction Scope: okSKILL.md and prompt.txt limit runtime behavior to producing text, short style notes, and optional evolution notes based on a provided reader_feedback field. There are no instructions to read system files, access environment variables, call external endpoints, or exfiltrate data.
Install Mechanism: okNo install spec or code files execute on disk; this is an instruction-only skill, which minimizes persistent or executable risk.
Credentials: okNo credentials, env vars, or config paths are required. The only input that affects behavior is reader_feedback (user-supplied text), which is appropriate for the stated feedback-driven evolution feature.
Persistence & Privilege: okalways is false and the skill does not request persistent system presence or modify other skills. It can be invoked autonomously by the agent (default behavior) but that is standard and not combined with other risky privileges here.