Crawler access (robots.txt)
2/2 · pass
No robots.txt (HTTP 404): every crawler is allowed by default
- GET /robots.txt -> HTTP 404
- Unmentioned (allowed by default): GPTBot, ClaudeBot, OAI-SearchBot, PerplexityBot, Google-Extended, Amazonbot, CCBot, Bytespider
Fix: Optional: publish a robots.txt that names the AI crawlers you allow and points to your sitemap; without one your policy is implicit.
Machine-readable surfaces
1/2 · warn
1 of 7 surfaces present (/openapi.json)
- /llms.txt: HTTP 404
- /sitemap.xml: HTTP 404
- /.well-known/x402: HTTP 404
- /.well-known/agent-card.json: HTTP 404
- /.well-known/mcp-registry-auth: HTTP 404
- /openapi.json: HTTP 200, valid JSON (OpenAPI 3.1.0)
- /ai.txt: HTTP 404
Fix: Publish /llms.txt (a short Markdown guide to your key pages); /sitemap.xml.
Readable without JavaScript
2/2 · pass
Home page is readable
- GET https://testgraph.21dle.co.uk/ -> HTTP 200, text/html, 17 KB
- 5,086 characters of visible text with scripts, styles and tags removed; <title>: "TestGraph — shared, evidence-backed memory for AI assistants"
- 0 external scripts, 0 inline; <noscript> absent
- first words: "TestGraph How it works Connect Account Live data Deliberations Example Try it API docs Experimental open-source infrastructure for AI Shared..."
Fix: Nothing to fix here.
Structured data
0/2 · fail
No JSON-LD, Open Graph incomplete
- 0 JSON-LD blocks
- Open Graph: no og:title; no og:description
- meta description: present
Fix: add a JSON-LD block (Organization or WebSite, plus Product, FAQPage or Article where relevant); add og:title and og:description so answer engines describe you accurately.
Legal notices (TAKE IT DOWN desk check)
0/2 · fail
0 of 6 required notices complete (desk score 0/12)
- Desk score 0/12 across 6 checks (1 of 12 pages read)
- Missing or partial: TAKE IT DOWN Act notice (0/2); DMCA / copyright notice (0/2); DSA notice-and-action (EU users) (0/2); Privacy request method (0/2); Accessibility statement and feedback (0/2); Marketing opt-out (0/2)
Fix: Publish the missing notices with a working request mechanism. Start with TAKE IT DOWN Act notice: Publish a plain-language notice and a request mechanism usable without an account; act within 48 hours.
MCP census entry (index.agentexchange.work)
1/2 · warn
Exposes /mcp but is not in the census
- Not in the census of 2,031 hosts selling to agents (2026-09-29)
- GET /mcp -> HTTP 405 (application/json): an endpoint answers
- Index MCP probe record: HTTP 404 (not probed by the index yet)
Fix: Publish /.well-known/x402 with your paid routes so payment indexers can list you.