Technical SEO & GEO — I make sites readable by Google and by the engines that answer instead of listing.
LinkedIn · angelhd029@gmail.com · Sevilla, Spain / remote
Español: SEO técnico y GEO. Hago que las webs se lean bien por Google y por los modelos que responden en vez de listar. Herramientas abajo; escríbeme en cualquiera de los dos idiomas.
Ranking and being cited are two different games. Site-level audits tell you whether the AI crawlers can get in; nothing told me whether a given page gave them anything worth quoting once they were inside — so I wrote the tools.
Nine of them, all Python, MIT, deterministic (no LLM calls, no API keys where avoidable), with tests and CI in green. This is the loop I run on real client sites.
All nine install with pip. Three of them — geo-check, seo-audit and render-gap — also ship as GitHub Actions, so a team can fail its own build the day a deploy blocks a citation bot or hides the copy behind JavaScript:
- uses: angelmunizpedraza/geo-check@v1
with: { url: https://example.com, fail-if-blocked: "true" }| Stage | Tool | What it answers |
|---|---|---|
| 1. Can the AI crawlers read me? | geo-check | Which engines can actually cite you, engine by engine — separating citation bots (OAI-SearchBot, PerplexityBot, Claude-SearchBot) from training bots (GPTBot, Google-Extended). Validates llms.txt, structured data and no-JS readability. |
| 2. Is anything there when they arrive? | render-gap | Compares the HTML your server returns with the DOM after JavaScript, and reports exactly which headings, paragraphs, links and JSON-LD only exist in a browser. Those crawlers do not run JS, so anything client-side is invisible to them. Scored 0-100 and CI-gateable. |
| 3. Give them a map | llms-txt-generator | Generates and validates llms.txt from the sitemap and the real content. |
| 4. Is this page quotable? | citeable | Scores one page 0-100 on 10 citability signals and returns an ordered list of which paragraph to rewrite. CI-gateable. |
| 5. Where do I rank but never get cited? | serp-to-ai-diff | Compares Google's top 10 with what ChatGPT and Perplexity actually cite, per keyword. That gap is the GEO backlog. |
| 6. Am I being cited at all? | ai-visibility-tracker | Brand mentions and citations inside AI answers, share of voice vs competitors, SQLite history, plus server-log analysis of real bot hits. |
| 7. Is my internal authority reaching the right pages? | linkjuice | Builds the internal link graph, finds orphans, click depth and internal PageRank, and returns concrete "link from this page to that page" recommendations. Counts body links only — nav and footer links are not votes. |
| 8. Is the classic SEO sound? | seo-audit | Crawls a site and returns indexing, content, performance and structured-data issues ranked by severity. |
| 9. Did it move traffic? | ga4-report | Organic traffic from the GA4 API with period-over-period comparison and landing-page alerts. Markdown, CSV or n8n webhook. |
- 4,000 → 28,000 organic visits/month (7×) and domain authority 12 → 27 on a group of online-education sites.
- +119 % organic traffic in 2 months and 10 keywords in the top 3 for a veterinary hospital in Los Angeles — English-language project, built end to end (React 19, Supabase, Cloudflare Workers).
- ~400 leads/month handled with no spreadsheets: n8n + Apps Script + CRM + WhatsApp Business API, first contact under 2 minutes.
Python · pytest · GitHub Actions · BeautifulSoup · SQLite · GA4 Data API · Search Console · WordPress / WooCommerce · React · Google Ads · Meta Ads · GTM · n8n · SQL
Technical SEO / GEO roles, remote or on-site in Seville. Also happy to receive issues and PRs on any of the tools above — feedback from a comment on LinkedIn is what produced geo-check v0.2, and that is the best kind.