Skillv1.0.0

ClawScan security

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

BenignMar 13, 2026, 11:20 AM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is an instruction-only SEO advisor whose declared purpose matches its runtime instructions and it requests no credentials or installs.
Guidance: This skill appears coherent and low-risk as provided. Before using it: do not provide site admin credentials or API keys unless you trust a follow-up request and understand why they are needed; be cautious if the agent asks to fetch private pages or run scripts on your site; verify any automated changes it proposes (avoid keyword-stuffing or copy-paste changes without review); if the skill later asks to install tools or access external services, treat that as a new, separate security decision.

Purpose & Capability: okThe name/description and the SKILL.md all describe keyword research, on-page and technical SEO advice; there are no unrelated environment variables, binaries, or installs requested.
Instruction Scope: okRuntime instructions are limited to SEO tasks (keyword suggestions, metadata, readability, technical recommendations). They do not instruct reading system files, using credentials, or sending data to external endpoints. Note: commands like "improve SEO for [URL/content]" imply the agent may fetch or analyze page content — that is expected for this purpose but should be performed only with user consent.
Install Mechanism: okNo install spec and no code files (instruction-only). This is the lowest-risk model: nothing will be written to disk or automatically installed.
Credentials: okThe skill declares no required environment variables, credentials, or config paths — consistent with an advice-only SEO tool.
Persistence & Privilege: okDefaults are used (not always:true, model invocation not disabled). The skill does not request elevated or persistent privileges beyond normal invocation.