site-visibility-report
timofeiryko/site-visibility-report
SEO / GEO / AEO website visibility audit -> client-ready HTML + PDF report. A reusable Claude skill (playbook + multi-agent measurement prompts + templates + scripts).
Instalação
npx convez-ai timofeiryko/site-visibility-report1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
README
A tradução para o português deste README ainda está na fila.
site-visibility-report
Repeatable SEO / GEO / AEO website audit → premium HTML + PDF report. The engine behind the ВИТРИНЫ
showcase reports (e.g. mobilexy.com). Measures how well a site is found in classic search, cited by AI engines,
and surfaced as direct answers — then renders a client-ready deliverable.
Quick start
- Open the report folder for the site (e.g.
ВИТРИНЫ/<domain>/). - Tell Claude: "Сделай site-visibility-report для https://" (или
/site-visibility-report). - Confirm language / measurement depth / structure (Step 0 in
SKILL.md). - Claude runs the pipeline and drops the HTML, standalone HTML, and PDF into the folder.
Manual (scripts)
pip install -r requirements.txt # requests, beautifulsoup4, lxml
# 1) measure — the bundled scripts/vendor/ are used automatically (no extra install)
python scripts/collect.py --domain https://example.com --pages "/,/pricing,/about,/blog" --out ./data
# multi-agent measurement prompts: prompts/recon-workflow.js + prompts/measure-workflow.js
# 2) score + fill templates/report-template.html (see SKILL.md + templates/STRUCTURE.md)
# 3) render
python scripts/render_pdf.py --html _pdf-source.html --pdf example-report.pdf --shot assets/_qa.png
python scripts/make_standalone.py --html example-report.html --out example-report.standalone.html
Files
SKILL.md— the playbook (read this first).prompts/— the multi-agent measurement orchestration (recon-workflow.js,measure-workflow.js).templates/report.css— design system.templates/report-template.html— 12-section skeleton.templates/STRUCTURE.md— section + scoring contract (HTML 12-sec, PDF 9-sec).scripts/collect.py— run measurement (uses the bundledscripts/vendor/by default).scripts/vendor/— bundled MIT scripts from geo-seo-claude (no separate install needed).scripts/make_standalone.py— inline CSS + images → one shareable file.scripts/render_pdf.py— headless-Chrome HTML→PDF + screenshot.report-data.example.json— worked data sheet documenting every field.requirements.txt·LICENSE(MIT) ·NOTICE(attribution).examples/— generate your own; rendered reports are git-ignored (seeexamples/README.md).
Credits & license
MIT (see LICENSE). Bundles MIT scripts from geo-seo-claude (© Zubair Trabzada) — see NOTICE +
scripts/vendor/LICENSE-geo-seo-claude.txt. Methodology & report format also draw on SEO-GEO-AEO-Skill
(Alex Labat — PDF format + 1–10 scoring), claude-seo (AgriciDaniel), marketingskills (Corey Haines).