Content Quality Auditor
PassAudited by ClawScan on May 10, 2026.
Overview
This is a coherent instruction-only content audit skill, but it may fetch web pages and save audit notes into persistent memory.
This skill appears safe for ordinary content-quality audits. Before using it on confidential or pre-release material, decide whether you are comfortable with WebFetch for URLs and with audit findings being saved into persistent memory files.
Findings (4)
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 spend time auditing newly created content automatically and may use the just-produced draft as audit input.
The skill may be invoked after content generation even when the user did not explicitly request an audit. This fits the content workflow, but it reduces the amount of setup confirmation.
PostToolUse hook auto-triggers: after content is written or substantially edited, the hook recommends this audit. When hook-triggered, skip setup questions — audit the content that was just produced.
Confirm whether hook-triggered audits are desired, especially for sensitive drafts or workflows where the user only wanted content creation.
URLs supplied for audit may be fetched from the web, and page content may be processed by the agent.
The skill can fetch web content, which is appropriate for auditing URLs and referenced benchmark materials.
allowed-tools: WebFetch
Only provide URLs or content that you are comfortable having the agent fetch and analyze.
Audit behavior or scoring guidance could change if the remote reference documents change.
The skill points to mutable GitHub 'main' branch reference documents that are outside the provided artifact set and may shape audit criteria or report format.
Full benchmark reference: [references/core-eeat-benchmark.md](https://github.com/aaron-he-zhu/seo-geo-claude-skills/blob/main/references/core-eeat-benchmark.md)
For reproducible or high-stakes audits, prefer pinned or bundled benchmark references.
Unpublished content issues or strategy notes may persist and influence future agent tasks.
The skill explicitly writes audit results into persistent memory paths without asking for confirmation.
Promotes: veto items and publish blockers to `memory/hot-cache.md` (auto-saved, no user confirmation needed). Top improvement priorities to `memory/open-loops.md`.
Review or disable auto-saving for sensitive drafts, and periodically clear or inspect the referenced memory files.