On-Page SEO Checker
Diagnose single-page SEO problems and turn them into prioritized, actionable fixes.
The skill limits its declared tool to WebFetch and explicitly treats fetched content as untrusted data. It lacks a complete authorization-confirmation, data-flow, and rollback treatment for target-site access and result handling, so points are deducted. No malware, credential theft, covert exfiltration, or destructive default is evident.
The procedure, input/output contract, N/A rules, and no-keyword fallback are reasonably coherent. However, the supplied files contain no skill-specific tests covering key paths, and the availability boundaries for WebFetch, local scripts, and Firecrawl are uncertain; the static ceiling therefore applies.
Triggers, scope boundaries, handoffs to E-E-A-T and technical SEO skills, single-page use, and bulk mode are described, with bilingual metadata. Actual trigger precision, page-type coverage, host compatibility, and regional reachability are not demonstrated, and Firecrawl may depend on an external service.
The documentation is layered through Quick Start, contract, data sources, numbered steps, templates, examples, and scoring guidance, and it declares Apache-2.0 and version 18.0.0. It does not clearly identify a maintainer, update path, or changelog for this skill, and SECURITY.md's supported 17.x line conflicts with the skill's 18.0.0 version.
The intended output, per-element scoring, prioritized repair plan, and handoff summary are directly specified, and the templates and example are useful. Still, the example is static author material rather than verified execution evidence, while missing metrics and tool data require user or manual review; the static ceiling limits the score.
The rubric, templates, evidence-labeling rules, and security boundaries provide some auditability. There are no skill-specific key-path tests, independent execution records, or corroborating third-party evidence, and the worked example does not establish real-world accuracy.
- This is a source-only review; WebFetch, scripts, and tests were not executed. Treat the worked example as illustrative, not verified output.
- Before fetching third-party sites, confirm authorization and comply with robots.txt, terms of service, and applicable law; the skill does not make authorization a fully specified decision gate.
- External fetchers such as Firecrawl may introduce data-egress and reachability concerns; mainland-China users should verify service availability and processing boundaries.
- The skill declares version 18.0.0 while SECURITY.md lists 17.x as the supported current line; reconcile release and maintenance status before adoption.
What does this skill do, and when should you use it?
This skill audits the page-level SEO health of a single URL and can investigate why one page lost rankings. It checks titles, meta descriptions, heading structure, keyword placement, links, images, page-level tags, and selected CORE-EEAT signals. The result is an evidence-backed scored diagnosis, a P0/P1/P2 repair plan, an overall score, and a handoff summary. It does not replace a full E-E-A-T publish-readiness audit or deep crawl, indexing, or Core Web Vitals analysis.
Reads a page URL or HTML, target keyword, page type, competitor URLs, and reported symptoms; uses WebFetch or optional local Python connectors to inspect title and meta data, H1/H2 structure, keyword placement, content structure, internal links, images, URL slug, canonical, and structured data; runs 11 audit steps, scores on-page elements out of 10, labels metrics as Measured, User-provided, or Estimated, and produces P0/P1/P2 recommendations, an overall score, and a handoff summary. If no target keyword is supplied, it infers candidates from the page and labels them Estimated.
- A content marketer wants to audit an article’s title, meta description, headings, and keyword placement.
- An SEO specialist needs to diagnose structural causes of a single page’s ranking drop.
- A publishing team wants a pre-launch check of page tags, links, images, and keyword coverage.
- A site owner wants to compare a page’s on-page SEO with a competitor URL.
- An SEO team needs a batch audit of five or more URLs with cluster-level findings and portfolio priorities.
What are this skill's strengths and limitations?
- Covers titles, meta descriptions, heading structure, keywords, links, images, and page-level tags.
- Scores elements out of 10 and ranks fixes as P0, P1, or P2.
- Separates measured, user-provided, and estimated metrics.
- Can proceed without a target keyword while clearly flagging the inferred keyword for confirmation.
- Apache-2.0 licensed and compatible with Claude Code and compatible Agent Skill hosts.
- Requires an accessible URL or usable HTML/content; it cannot audit without page input.
- Optional search-volume and competitor metrics may be unavailable and are reported as N/A.
- CORE-EEAT is only a quick scan here, not a full publish-readiness verdict.
- Deep crawl, indexing, Core Web Vitals, mobile rendering, and HTTPS/security issues are out of scope and routed to technical-seo-checker.
- The source provides no independent test results or platform coverage matrix for this individual skill.
How do you install this skill?
Get the repository with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. Compatible hosts can use npx skills add aaron-he-zhu/aaron-marketing-skills -s on-page-seo-checker. The source does not document a separate directory-placement procedure for other hosts.
How do you use this skill?
Example trigger: Audit the on-page SEO of https://example.com/page. With a keyword: Check SEO issues on this page targeting [keyword]: [URL/content]. For comparison: Compare on-page SEO of [your URL] vs [competitor URL] for [keyword]. Provide a URL or page content; if neither is available or inferable, the skill asks for one.
How does this skill compare with similar options?
Boundary with content-quality-auditor: this skill evaluates page structure and page-level SEO signals, while content-quality-auditor performs the full CORE-EEAT publish-readiness audit. Boundary with technical-seo-checker: this skill does not handle deep crawlability, indexing, Core Web Vitals, or technical SEO diagnostics.