On Page Seo Auditor
Security checks across malware telemetry and agentic risk
Overview
This instruction-only SEO auditor is coherent and generally safe, but users should notice that it can fetch webpages, optionally use connected SEO/Search Console data, and write reusable audit summaries to memory.
This appears reasonable to install for on-page SEO audits. Before using it, confirm bulk URL scope, avoid sharing pages or account data you do not want analyzed, be careful when connected Search Console or SEO-tool accounts are available, and review any memory/audit summaries before keeping them for future use.
VirusTotal
65/65 vendors flagged this skill as clean.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may retrieve one or more webpages during an audit, especially for competitor or bulk comparisons.
The skill is allowed to fetch web pages, which is central to auditing SEO for supplied URLs but still means the agent will access external web content.
allowed-tools: WebFetch
Provide only URLs you are comfortable having the agent fetch, and confirm the page list or sampling approach before large audits.
If such integrations are connected, the audit could use private site analytics or search performance information.
The skill may use connected SEO or Search Console integrations if present; this is relevant to SEO auditing, but those services can contain account- or site-specific performance data.
Use ~~web crawler, ~~SEO tool, and ~~search console when connected; otherwise ask for page URL/HTML, target keywords, and competitor URLs.
Use least-privilege connections and make sure the agent is using only the intended site property or SEO tool account.
SEO findings, priorities, and possibly competitor or business context may be retained for later tasks.
The skill creates persistent audit summaries and open-loop items, which can be useful but may influence future agent work or retain business/site details.
Writes: a user-facing audit or optimization plan plus a reusable summary that can be stored under `memory/audits/`. Promotes: blocking defects, repeated weaknesses, fix priorities, and pending decisions to `memory/open-loops.md`.
Review memory entries before saving them and avoid storing sensitive business details that are not needed for future SEO work.
If the agent relies on remote referenced documents, behavior may depend on content outside the packaged skill review.
Some workflow and policy references point to external GitHub documents that were not included in the reviewed artifact set, so those referenced instructions were not verified here.
finish with the standard handoff summary from [Skill Contract](https://github.com/aaron-he-zhu/seo-geo-claude-skills/blob/main/references/skill-contract.md)
Prefer bundled or pinned reference documents for repeatable behavior, and review any external linked instructions before relying on them.