Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
DeepSeek
Muse+22 moreClaude CodeCursorCodex CLIDeepSeekMuse
Dots
Manus
Grok
Grok Bot+18 agentsSite Visibility Diagnostic
Turn a full-site crawl into a scored SEO visibility audit with a ranked, owner-ready remediation plan.
$5
Site Visibility Diagnostic
Example session with this skill installed
Our lead-gen site https://www.acme-consulting.com is down about 30% on organic since a redesign last March and nobody can tell me why. Please run a full SEO health check across the site: homepage, services, case studies, blog. Skip our /wp-admin and client login areas. It is roughly 400 pages. Check robots.txt, sitemap, canonicals, redirects, schema, and mobile performance. I want a visibility score, an ordered list of issues with severity, and a two-week fix plan that tells me what to hand to my developer versus what I can fix myself in the CMS.
- Read your context and instructions
- Compiled the site visibility diagnostic
Summary
Visibility score
48/100 (high confidence) — mixed catalog + lead capture: service pages plus editorial blog. 312 pages reviewed, ~78% of sitemap URLs covered. One-line diagnosis: noindex on the pricing template and 5-hop redirect chains on category hubs are suppressing indexing of revenue pages.
| Lane | Weight | Score | Verdict |
|---|---|---|---|
| Discoverability & indexability | 25% | 35 | Critical: 14 pricing pages noindexed via meta robots (confirmed) |
| Content usefulness & trust | 25% | 55 | Likely: case studies lack author or client attribution |
| Template on-page quality | 20% | 60 | High: 203 duplicate titles across the service template |
| Structured data & entities | 10% | 40 | Missing Service and BreadcrumbList JSON-LD on key pages |
| Performance & UX | 10% | 52 | LCP risk: 2.1 MB unoptimized hero JPEG, no srcset |
Next steps
- Week 1: remove
noindexfrom the /pricing/* template and resubmit the sitemap — dev, ~1h - Week 1: flatten category redirect chains to single 301s — dev, ~2h
- Week 2: unique titles/H1 per service page; add author and client attribution to case studies — content
- Week 2: deploy Service and Breadcrumb JSON-LD, then re-crawl to verify
- Full artifacts:
visibility-diagnostic-report.md,remediation-roadmap.md,evidence-summary.json
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Vague SEO reports often fail to distinguish between cosmetic issues and critical indexation blockers. Developers and site owners struggle to turn raw crawl data into a prioritized, actionable technical roadmap.
What it does
- Performs a staged diagnostic crawl of up to 500 pages while respecting robots.txt and crawl budgets.
- Identifies site-type models to apply specific evaluation logic for e-commerce, SaaS, lead gen, or editorial sites.
- Analyzes technical infrastructure, content credibility, template-level health, and entity clarity via structured data.
- Generates a weighted visibility score from 0-100 based on discoverability, UX, and AI extractability.
- Produces a tiered remediation plan categorized by severity, effort, and ownership.
Why this beats prompting it yourself
General prompts often result in repetitive lists of generic SEO advice. This skill enforces a multi-lane diagnostic framework that separates infrastructure risks from content quality, ensuring every finding is backed by specific URL evidence and template-level sampling.
Use cases
- Conducting a pre-launch technical SEO audit for a production environment.
- Diagnosing why specific site sections are not appearing in search results.
- Benchmarking a site's readiness for AI-assisted discovery and answer engines.
- Creating a developer-ready backlog for technical SEO debt remediation.
Known limitations
Default crawl limit is 500 pages. Visual QA requires screenshot capabilities. Backlink prospecting and paid search analysis are not supported.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills to Reverse Engineer and Scrape Websites.
Core Web Vitals are inferred from page weight and markup when live measurement is unavailable, so some performance findings are labeled Likely rather than Confirmed. Sites far above the 500-page budget get template-level coverage, not exhaustive page-level review.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
This skill has been used 5 times.
Trust & safety
Security scan
Security scan passed 1 month ago
Permissions
File Scopes
Network and browser access are needed to fetch, crawl and screenshot the user's own public site, and local read/write for the audit artifacts; the crawler stays on the audited domain and respects robots.txt.