Skillv1.0.0

ClawScan security

Human Writing · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 1, 2026, 5:10 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only writing-style assistant that is internally consistent with its stated purpose and does not request credentials, install software, or contact external endpoints.
Guidance: This skill appears coherent and low-risk: it’s instruction-only, uses an included style checklist, and asks for no credentials or installs. Before installing, consider that the source/homepage is missing and the owner is unknown — if provenance matters to you, prefer skills with a known publisher. Also remember the skill enforces a particular tone and set of edits; review any changes before publishing (and avoid feeding sensitive secrets into drafts). Finally, absence of scan findings only means there were no code files to scan — it’s still good practice to inspect the SKILL.md and bundled reference (which you can) before use.

Purpose & Capability: okThe skill's name and description (produce natural, human-sounding copy) match the contents of SKILL.md and the included anti-AI patterns reference. It does not request unrelated binaries, credentials, or config paths.
Instruction Scope: okRuntime instructions are limited to applying style rules and scanning the draft against the bundled references/anti-ai-patterns.md file (explicitly referenced). There are no instructions to read arbitrary system files, access environment variables, or send data to third-party endpoints.
Install Mechanism: okNo install spec and no code files — instruction-only. This minimizes disk writes and execution of external code.
Credentials: okThe skill declares no required environment variables, credentials, or config paths. The requested access (none) is proportional to the stated writing/editing purpose.
Persistence & Privilege: okalways is false and the skill does not request system-wide changes or modify other skills. Normal autonomous invocation is allowed (platform default) but not excessive for this type of skill.