
GEOlikeaPro
Find out where ChatGPT, Claude & Perplexity skip your brand
2 followers
Find out where ChatGPT, Claude & Perplexity skip your brand
2 followers
Find out if ChatGPT, Perplexity and Gemini recommend your store. GEOlikeaPro measures your AI Share of Voice and shows you exactly what to fix. Start free.








π Big update β Agent Standards Validator is now real.
The old version was a single LLM call asking the model to guess whether a site implemented MCP / A2A / UCP based on news articles. We rewrote it from the ground up as a deterministic probe rig:
- 16 `.well-known/*` paths, fired in parallel
- Real MCP `initialize` JSON-RPC 2.0 handshake + `tools/list`
- A2A agent-card schema validation (8 required fields)
- 20-UA AI crawler audit β flags the "robots.txt-says-allowed-but-edge-403s" class of bug
- Wikidata anchor via SPARQL on official-website URL
- Wayback Machine fallback when Cloudflare bot management blocks the live edge
- Copy-pasteable starter files for every gap (`agent-card.json`, `llms.txt`, `mcp-server.ts` with stack-aware mount notes, agentic-commerce.json, plus a Wikidata submission brief)
Free to run at the Agent Standards tab.
Curious which verdict feels wrong on your site β please ping me here if you spot anything off π
Hey Product Hunt,
Quick problem: most "is my site AI-friendly?" checkers spoof a user-agent header from your edge and call it a day. That's useless. Hosts like Hostinger shared, default-on Cloudflare, Imperva, Sucuri block AI crawlers by IP range, not user-agent. Spoofing GPTBot from your laptop passes through; the actual GPTBot from OpenAI's datacenter gets a reCAPTCHA wall.
Hosting Checker takes a different approach. For each URL, it runs three layers in parallel:
Real fetch via the AI itself β we ask ChatGPT Search and Perplexity to fetch your URL with their own browsing tools, from their own IPs. Ground truth, not spoofing. If they get a challenge page, they report exactly what they saw.
Origin fingerprint β ASN lookup against known-problematic hosts (Hostinger AS47583/AS214572, Cloudflare default-on, Imperva, Sucuri, Akamai), plus response-header signature scanning.
robots.txt parse β per-bot allow/disallow for the 8 AI crawlers that actually matter today: GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-User, PerplexityBot, Google-Extended, Applebot-Extended.
Output: an Access Score, per-bot verdicts (with the actual error string when blocked), and a callout panel naming your host's known issues.
Free tier: 50 units/month, ~16 full scans. Bring your own OpenAI key for unlimited.
What I'd love feedback on: which other bot-protection vendors should we fingerprint? Right now we cover the top 5 β what's silently breaking your AI visibility?
Biggest UX pass since launch. Built side-by-side with an enterprise SEO who uses the tool every day.
Home cockpit replaces the welcome card. Latest audit SOV / models cited / avg rank / GVI up top, plus a Next-Best-Action card that triage-routes you to whatever needs fixing first.
Cmd+K command palette on desktop. Type a few letters, hit Enter, jump anywhere β anything you can reach via the sidebar, you can reach faster via the keyboard.
Mobile native share sheet. Two Export buttons replaced by one "β Share" that opens the OS share sheet β Mail, Drive, WhatsApp, Save to Files, AirDrop. No more sideways scrolling on a phone to find Export PDF.
Audit unified. SOV / Models / Recommendations / Vitals were four peer tabs in the sidebar. Now one workspace with an inner sub-nav strip β the workflow finally reads as one thing.
Cleaner visual language. Mono `// section` comment labels, unified geometric glyphs, accent-tint active states. Distinct from the 500+ other GEO / AI-search tools.
Live at geolikeapro.com/tool β feedback welcome.