Agent Discoverability Score
Live URL-input scan across the technical dimensions of agent discoverability. Public score on screen — no signup. Slice 2a covers 4 of 6 dimensions today (llms.txt, MCP, Structured Data, Agent-Parsable Content); the remaining two ship in upcoming releases.
Want the deeper analysis now? The $79 Audit delivers a full report with live citation data in 60 seconds. Run your audit →
The six dimensions
Each dimension maps to a specific piece of technical infrastructure. The score tells you where you stand; the gap report tells you exactly what to fix.
llms.txt Quality
15%Machine-readable sitemap telling agents what your site offers, and where.
MCP Server Discoverability
20%Your agent-invocable endpoint, discoverable via .well-known/mcp.json.
OpenAPI / API Catalog
10%Structured API description for service-based businesses.
Structured Capability Data
20%JSON-LD schema: Organization, Service, Offer, FAQPage.
Agent-Parsable Content
15%Pages that render cleanly without JavaScript; pricing in text, not images.
Citation Visibility & Monitoring
20%Live audit of where you're cited across ChatGPT, Claude, Perplexity, Gemini.
Why this, not the free Cloudflare tool?
Cloudflare shipped a free Agent Readiness Score tool in 2026 that does pass/fail checks across five categories. It's a good free check. We go deeper in three places:
Quality rubrics, not pass/fail
Our llms.txt check doesn't just verify the file exists; it scores curation quality, blockquote elevator-pitch, and whether the linked pages return good markdown. Same depth on every other dimension.
Predicted referral lift per gap
Each gap is annotated with an estimated impact on agent-attributed traffic. You fix what moves the needle first, not what's easy to check off.
Live citation audit
We query ChatGPT, Claude, Perplexity, and Gemini for prompts in your category and report your current citation share against competitors. Cloudflare can't see inside those engines. We can.
Built the way we build for clients
The scanner runs on Cloudflare Workers. Citation audits go through Profound's enterprise API. Everything is MCP-callable — agents can invoke the scanner against any URL programmatically, the same way our clients' MCPs are invoked. When the score ships, we'll score ourselves first and publish the result here.
See our MCP server → https://mcp.astrant.io/mcp
FAQ
- What's the difference between Score and Audit?
- The free Score gives you a public grade across the dimensions we currently cover. The $79 Audit adds live citation audit across major AI engines, competitor comparison, implementation estimates, and a JSON export for programmatic use. If you just want to know where you stand, use the Score. If you want a prioritized action plan, use the Audit.
- Why does it say "4 of 6 dimensions"?
- Slice 2a ships four dimensions: llms.txt Quality, MCP Server Discoverability, Structured Capability Data, and Agent-Parsable Content. The remaining two — OpenAPI Documentation (Dim 3) and Citation Visibility (Dim 6) — ship in upcoming releases. Your score is computed on what's currently scored; when more dimensions ship, your scan will rerun automatically if you opted into the monthly rescan.
- What's the free tier vs paid tier difference for Dim 5?
- Dim 5 (Agent-Parsable Content) checks whether your homepage content is visible without JavaScript. The free Score uses a static fetch — it tells you if the static-only signal is healthy. The $79 Audit additionally runs a real browser render and diffs the two: the diff is what reveals which content (pricing, FAQs, capabilities) requires JS to appear. If your site is static-rendered or SSR'd, the free check is sufficient. If it's a JS-heavy SPA, the Audit's render diff is where the real signal lives.
- Will you charge for the Score later?
- No. The Score stays free. The paid tiers (Audit, Implementation, Custom, AutoPilot, Concierge) go deeper.
Want a deeper read?
The free Score tells you where you stand. The $79 Audit gives you a prioritized action plan with live citation data and competitor comparison — delivered as a PDF in 60 seconds.