Skillv1.0.2

ClawScan security

SEO GEO for SaaS · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 5, 2026, 6:43 PM

Verdict: benign
Confidence: medium
Model: gpt-5-mini
Summary: The skill's requirements and runtime instructions are coherent for an SEO tool: the Google and DataForSEO credentials and use of curl align with the stated tasks, but because the source is unknown and the skill asks for sensitive API credentials you should exercise caution before granting access.
Guidance: This skill appears internally consistent for SEO work, but because it asks for sensitive API credentials and the package has no public homepage or identifiable author you should take precautions before installing: 1) Prefer granting read-only, least-privilege Google access (or use a service account / limited OAuth scopes) rather than full account credentials or a long-lived refresh token if possible. 2) Store secrets in a secrets manager rather than pasting them into chat. 3) Back up and review any existing seo/ files before running the onboarding flow — the skill will create/modify files in your project. 4) Review generated content and any quoted/external-sourced material for accuracy and copyright concerns before publishing. 5) Verify and inspect the recommended companion skill (seo-geo / dataforseo / ga4) implementations before invoking them. 6) If you are unsure about granting tokens to an unknown source, run onboarding manually (export GSC/Ga4/DataForSEO CSVs) instead of providing live credentials.

Review Dimensions

Purpose & Capability: okThe skill is an SEO/GEO system and it requests Google OAuth credentials (client id/secret/refresh token) and DataForSEO login — both are reasonable for pulling Search Console/GA4 and DataForSEO data. curl is used for robots.txt and bot-access checks and matches the documented workflows. No unrelated services, binaries, or config paths are requested.
Instruction Scope: okSKILL.md is instruction-only and stays within SEO scope: it reads/writes an seo/ directory in the user's project (creates overview/keywords/opportunities/calendar/etc), pulls data from Search Console / GA4 / DataForSEO / SemRush (when available), and runs pre-publish checks (robots.txt, schema validation). There are no instructions to read unrelated system files or user shell history. Note: the skill will create and modify files in the user's project — review those outputs before publishing.
Install Mechanism: okNo install spec or external downloads — instruction-only. This minimizes risk because no archive or arbitrary executable is fetched or written to disk by an installer step.
Credentials: noteThe required env vars (GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, GOOGLE_REFRESH_TOKEN, DATAFORSEO_LOGIN, DATAFORSEO_PASSWORD) are proportionate to the described integrations. However these are sensitive credentials: Google tokens give access to Search Console/GA4 data and DataForSEO credentials give API access. The skill asks for multiple secrets, which is expected for this use case but warrants careful handling (use least-privilege tokens, read-only scopes, service accounts where possible).
Persistence & Privilege: okThe skill does not request always:true and does not try to modify other skills or system-wide configs. Its runtime behavior writes files into the user's project (creates seo/ and reference files) which is expected for its purpose. Autonomous model invocation is allowed by default but not, by itself, a red flag here.