Audit SEO and AEO Content Quality with Scored Reports

Audits a URL, website, codebase, or pasted content for SEO and AEO (answer-engine optimization) performance, triggered by phrases like "audit this content", "SEO review", "AEO audit", "score this article", or "analyze content quality". Input is a structured object naming the content or path, its input_type (text, url, website, or codebase), primary and secondary keywords, content type (landing page, blog post, pillar page, product page), a word-count target defaulting to 1500, optional business context and conversion goal, and existing published content. For a new project it defaults to a full site/codebase audit before recommending new content; if target keywords are missing, it identifies candidates from the site and asks which matter before finalizing strategy, falling back to a clearly labelled, observed-search-intent-based strategy if the user doesn't know. Audit dimensions, evidence rules and the audit-to-implementation loop live in references/seo-audit-checklist.md.

Every score is out of 100 against a fixed threshold table: 85-100 is a passing "Strong" score needing minor polish, 70-84 is an acceptable "Warn" needing recommended fixes, 50-69 is a "Weak" score needing work before publishing, and 0-49 is a "Fail" that should not publish as-is. Output always follows one fixed Markdown report structure: a summary table of Overall/SEO/AEO/Readability scores with a one-sentence verdict naming the single most important fix; an SEO report scoring passing checks, critical issues (each with a specific problem, severity, and exact fix instruction), warnings, a keyword-occurrence/density table against target and status, a heading-structure table checking H1 count (must be exactly 1), H2 count (recommended 4-6 for blog posts) and H3 nesting, and a meta-elements table checking title tag (50-60 characters), meta description (140-160 characters) and keyword presence; an AEO report checking for a TL;DR/direct-answer block immediately after the H1, definition sentences, an FAQ section (minimum 4 questions), numbered/bullet lists (minimum 2), comparison tables, primary keyword in the first 100 words, concise under-50-word answers, and schema markup, plus an extractability table rating whether AI engines can likely extract answers to "What is [topic]?", "How does [topic] work?", "How much does [topic] cost?" and "What are the benefits?" style questions; a readability report checking passive-voice percentage (under 10%), transition-word presence, wall-of-text paragraphs (target zero), subheading frequency (every ~300 words), and reading level (grade 7-9); a prioritized fix list split into Do First (critical, blocks publishing), Do Second (important, affects ranking), and Do Last (polish); and a projected-score table comparing current to post-fix scores for SEO, AEO, readability and overall.

Execution runs in ten steps: detect the input type and inspect the codebase, fetch the URL, or analyze the supplied text; establish business context, audience, conversion goal, existing content and target keywords, asking for confirmation on any candidate keywords before treating them as final; check technical SEO (crawlability, indexability, robots.txt, sitemap, canonical URLs, redirects, status codes, metadata, structured data, mobile/performance signals, image alt text, broken links); check information architecture and content (page purpose, search intent, heading hierarchy, topical coverage, duplication, cannibalization, thin pages, internal links, orphan pages, conversion paths); run keyword analysis reporting counts as evidence rather than a mandatory density target; run the AEO signal checklist and extractability assessment; score SEO, AEO, readability and conversion readiness with confidence and evidence per finding; build a prioritized fix plan naming exact files/pages, implementation notes, dependencies and verification criteria; if authorized, apply fixes in priority order and re-run the audit to report before/after results, never publishing or mutating external accounts without explicit authorization; and verify findings against the actual supplied content or project files, using any repository audit helper only as a supplementary check rather than a substitute for evidence-based review or an unsupported density score.

Use it before implementing SEO/AEO changes on a website, codebase, URL, page or content set, and again afterward to verify the fixes. Skip it when the goal is proving indexing, ranking, conversion or third-party platform state, since an audit cannot confirm those without current observable evidence.

Input is the content, URL, website, or codebase to audit, plus target keywords, content type, word-count target and optional business context. Output is the fixed-format Content Quality Audit Report: summary scorecard, SEO/AEO/readability sub-reports with evidence-backed findings, a prioritized fix list, and projected post-fix scores.

Receives input from the seo-aeo-blog-writer and seo-aeo-landing-page-writer skills and feeds its findings to seo-aeo-internal-linking and seo-aeo-schema-generator; reads its audit dimensions and evidence rules from references/seo-audit-checklist.md.

Content and SEO teams who need an evidence-based, consistently scored SEO/AEO/readability audit of a page, article, or whole site before publishing or as a verification pass after fixes.