SERP Analysis
Assess whether a query is realistically targetable by mapping its SERP structure, ranking patterns, and feature opportunities.
The skill explicitly treats WebFetch content as untrusted data and points to robots.txt, scraping boundaries, and data-egress safeguards; the repository also states that connectors are read-only by default and do not persist keys. Points are deducted because memory writes and promotion to persistent state lack per-action user confirmation, rollback guidance, and a complete data-flow disclosure, while Firecrawl/Tavily can expose queries and target URLs to third parties.
The ordered method, input/output contract, missing-tool fallback, and metric-labeling rules are reasonably consistent; repository CI, tests, and zero-dependency connectors provide supporting signals. Static review cannot reproduce live SERP retrieval, citation, or memory-write paths, and the declared WebFetch tool does not fully align with the documented Python connector paths, so the score is capped below 10 and reduced further for limited abnormal-input and failure-feedback coverage.
Trigger wording, the non-fit boundary for keyword-demand discovery, required inputs, expected outputs, and multi-keyword, geographic, and device variants are specified, with bilingual naming. Points are deducted because search-engine differences, Chinese SERP behavior, and mainland-China network reachability are not validated, while AI Overview and related platform capabilities depend on live external services.
Frontmatter supplies naming, version, license, compatibility, argument hint, and next-skill metadata; the referenced taxonomy, templates, example, and shared contract provide substantial structure, and Apache-2.0 plus repository CI/maintenance signals are present. Points are deducted because the skill lacks its own changelog, named maintenance owner, or explicit update path; its references and commands depend on the host repository layout, and example metrics lack source attribution.
The intended result is a usable SERP brief containing intent, difficulty, ranking patterns, feature strategy, and recommendations, and the templates and example make the output shape fairly direct. Points are deducted because core conclusions require a live or user-supplied SERP and no real execution artifact is provided; authority, link, and stability inputs may be unavailable, leaving substantial manual verification, so the static score cannot exceed 7.
The skill requires every metric to be labeled Measured, User-provided, or Estimated and requires SERP evidence for difficulty and intent claims; the repository also provides security documentation, CI workflows, and relevant tests. Points are deducted because no independently reproducible execution report, SERP sample, or third-party corroboration for this skill is included, and the example's figures and conclusions cannot be independently verified from the supplied files, so the score is capped below 5.
- Firecrawl or Tavily may send queries and target URLs to third parties; do not submit confidential queries or crawl sites without authorization.
- Do not treat the example's DA, backlink, difficulty, or AI Overview figures as measured data; reacquire and label evidence before relying on them.
- The skill recommends writing to memory, but the supplied skill text does not define per-write confirmation, recovery, or deletion procedures; confirm scope manually and preserve recoverable state.
- Validate SERPs separately for mainland-China location, language, and device conditions because the supplied material does not establish reliable reachability of live search or overseas services.
What does this skill do, and when should you use it?
SERP Analysis is the Survey-phase SEO/GEO skill in Aaron Marketing Skills. It analyzes search-result composition, search intent, ranking factors, AI Overviews, featured snippets, People Also Ask, and other SERP modules for a query. It reads the keyword, location and language, device, screenshots or top-10 URLs, and search context, then produces a prioritized SERP brief. The skill also requires an evidence-backed True Difficulty score from 0–100 and fit guidance for new, growing, and established sites.
Reads target keywords, location and language, device, SERP screenshots, top-10 URLs, and search context; maps visible ads, AI Overviews, featured snippets, People Also Ask, knowledge panels, image or video packs, local packs, shopping, news, sitelinks, and related searches; analyzes ranking-page URLs, authority, format, freshness, on-page factors, and structure; identifies ranking patterns, dominant intent, and feature opportunities; calculates True Difficulty using the specified weighted inputs; and produces key findings, minimum content requirements, SERP feature strategy, a recommended outline, next steps, and a reusable research handoff summary.
- An SEO team needs to validate the dominant intent and competitive bar for a target query.
- A content lead is planning an article, landing page, or product page against the current SERP.
- A GEO team is assessing opportunities in AI Overviews, featured snippets, or People Also Ask.
- A local or international team needs to compare SERPs by location, language, or device.
- A strategist wants to compare several query SERPs or investigate historical SERP changes.
What are this skill's strengths and limitations?
- Covers SERP composition, ranking factors, intent, and AI-search features rather than only keyword lists.
- Requires difficulty and intent claims to be supported by live or provided SERP evidence.
- Produces a 0–100 True Difficulty score, site-stage guidance, and a standardized handoff.
- Supports multi-keyword, historical, local, and mobile-versus-desktop comparisons.
- SERP feature composition requires a hand-checked screenshot; Firecrawl or Tavily organic-result lists alone cannot establish ads, AI Overviews, PAA, or other modules.
- Live analysis depends on WebFetch or an available network connector; otherwise the user must supply the evidence.
- The supplied source does not show an independent test suite or platform-by-platform verification for this skill.
- It is not for keyword-demand discovery; the repository directs that work to keyword-research.
How do you install this skill?
For the full repository, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills and then /plugin install aaron-marketing@aaron in Claude Code. On an Agent Skills-compatible host, run npx skills add aaron-he-zhu/aaron-marketing-skills, or clone it with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The repository supports an npx skills add ... -s form for a single skill, but the supplied source does not show the complete command specifically for serp-analysis.
How do you use this skill?
Use a prompt such as Analyze the SERP for [keyword] or What does it take to rank for [keyword]? Provide the target keyword, location and language, device, and either a SERP screenshot or top-10 URLs. Without tools, the skill asks for those inputs. Label every metric as Measured, User-provided, or Estimated, and mark unavailable required metrics as N/A.
How does this skill compare with similar options?
Unlike the repository's keyword-research skill, SERP Analysis studies the current result page, ranking patterns, intent, and feature opportunities for selected queries rather than discovering keyword demand.