# Origin Website Audit > Comprehensive audit for landing pages and websites across advertising and legal compliance, SEO, marketing conversion quality, design/UX, accessibility, technical health, and analytics/measurement. Use when asked to review, audit, evaluate, QA, or preflight a website, LP, service site, corporate site, article site, recruiting site, or form-based conversion flow before or after launch. - Skill: `aki2022/origin-website-audit` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add aki2022/origin-website-audit` - Raw SKILL.md: https://api.skillmd.com/api/skills/aki2022/origin-website-audit/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Marketing & Growth - Author: Aki2022 (https://skillmd.com/u/aki2022) - Updated: 2026-09-10 - Page: https://skillmd.com/skills/aki2022/origin-website-audit --- # Website Audit Use this skill to audit websites and landing pages comprehensively. The goal is to find launch-blocking and performance-limiting issues across compliance, marketing, SEO, UX, accessibility, technical quality, and measurement. ## Inputs Accept any of these targets: - Public URL, staging URL, or local dev server URL - Static HTML or built site files - Source files for a page, app route, component, or template - Screenshots when browser access is unavailable When provided, also read only the relevant context files: - PRD, campaign brief, ad copy, keyword list, target audience, or conversion goal - Brand/design guidelines - Legal, privacy, pricing, offer, or industry-specific requirements - Ad platform requirements such as Google Ads policy notes - Existing analytics/GTM/ads conversion specifications If context is missing, continue with generic website best practices and mark assumptions clearly. ## Core Workflow 1. Identify the site type, primary audience, main conversion goal, traffic sources, and target pages. 2. Inspect the rendered experience when a URL is available. Test mobile, tablet, desktop, and wide desktop where practical. 3. Read source files only as needed to validate findings or propose concrete fixes. 4. Evaluate every required category below. 5. Report findings by severity with evidence, impact, recommended fix, and verification method. 6. Separate confirmed issues from assumptions or items needing human/legal review. ## Required Audit Categories ### Advertising and Legal Compliance Check for risks that could cause ad disapproval, user deception, or legal exposure: - Platform policy fit for Google Ads and other declared ad platforms - Prohibited or restricted category claims based on the site topic - Exaggerated, absolute, fear-based, or unverifiable claims - Unsupported performance numbers, rankings, testimonials, guarantees, or before/after claims - Misleading redirects, disguised ads, missing destination relevance, or mismatch with ad copy - Clear operator/business identity, contact path, privacy policy, terms, pricing, conditions, disclaimers, and eligibility limits - Proper handling of sensitive categories such as health, finance, employment, housing, legal, adult, politics, and personal hardship Do not present legal conclusions as legal advice. Flag legal-sensitive items as review risks when a qualified review is needed. ### Marketing and Conversion Quality Check whether the page can convert the intended traffic: - Match between traffic intent/ad copy/keywords and above-the-fold message - Clear offer, audience fit, differentiation, and trust proof - CTA visibility, specificity, repetition, and continuity through the page - Form length, field friction, validation clarity, confirmation state, and fallback contact path - Objection handling, social proof, process explanation, pricing clarity, and next-step expectations - Mobile conversion path and tap target ergonomics ### SEO Check discoverability and search intent alignment: - Title, meta description, canonical, robots, sitemap, indexability, and URL hygiene - H1/H2 structure, semantic sectioning, internal links, breadcrumbs where relevant - Search intent match, topical coverage, originality, and content depth - E-E-A-T signals: author/operator clarity, sources, dates, credentials, reviews, and trust pages - Structured data opportunities and implementation risks - Duplicate/thin content, hidden text, doorway-page patterns, and keyword stuffing ### Design and UX Check visual quality and usability: - Responsive behavior at mobile, tablet, desktop, and wide desktop - Layout overflow, overlap, clipping, awkward wrapping, and unstable spacing - Visual hierarchy, scanability, CTA prominence, navigation clarity, and wayfinding - Typography, color, spacing, component consistency, and brand fit - Form experience, error recovery, loading/empty states, and confirmation flows - Avoid generic feedback; tie design findings to user comprehension, trust, or conversion impact When available, use `web-design-reviewer` for visual/responsive inspection. If it is not available, use `web-design-guidelines`, `playwright-cli`, browser automation, screenshots, or source review as practical. ### Accessibility Check the highest-impact WCAG-style issues: - Accessible names for links, buttons, controls, icons, and form fields - Keyboard access, visible focus, focus order, and dialog/flyout behavior - Semantic HTML, heading order, landmark structure, lists, and tables - Color contrast, non-color indicators, reduced motion, and readable text sizing - Image alt text, media captions where relevant, and form error association Use `fixing-accessibility` when detailed remediation guidance is needed. ### Technical and Measurement Quality Check whether the site can be trusted, measured, and operated: - Console errors, broken requests, broken links, redirects, 404s, mixed content, and SSL issues - Core Web Vitals and performance risks: LCP, CLS, INP, image sizing, render-blocking resources, third-party scripts - OGP/Twitter cards, favicon, manifest, localization tags where relevant - GA4/GTM/ad conversion events, event naming, consent behavior, duplicate firing, and thank-you/lead state tracking - Error monitoring, form submission reliability, spam protection, and data privacy boundaries For React or Next.js code, use `vercel-react-best-practices` when performance, rendering, bundle size, or data-fetching risks are in scope. ## Severity Rubric - `P0` Launch/ad-blocking: likely ad disapproval, serious legal/privacy risk, broken conversion path, severe accessibility blocker, or site unavailable. - `P1` High impact: materially hurts conversion, SEO indexability, trust, measurement accuracy, or core mobile UX. - `P2` Medium impact: weakens clarity, performance, accessibility, maintainability, or marketing effectiveness but does not block launch. - `P3` Low impact: polish, minor consistency, optional SEO/UX enhancement, or future optimization. ## Output Format Use this structure: ```markdown # Website Audit Report ## Summary - Target: - Site type: - Primary goal: - Overall status: - Top risks: <3-5 concise bullets> ## Findings ### [P0|P1|P2|P3] - Category: - Evidence: - Impact: - Recommendation: - Verification: ## Positive Notes - ## Assumptions and Review Needed - ## Suggested Next Checks - ``` Keep the report evidence-based. Do not invent policy failures or metrics that were not observed. If browsing, live testing, or source access is unavailable, state the limitation and audit the available artifact. ## External Skill Notes - Prefer installed, trusted local skills over low-install external skills. - `web-design-reviewer` is a recommended optional helper for design and responsive inspection if installed. - Low-install advertising or LP audit skills may be used only as reference material after reading their instructions; do not rely on them as authoritative policy sources. - For up-to-date ad platform policy, official platform documentation should be checked when the decision affects launch, spend, or account risk.