belgravialabs.com agentic web readiness
A+ Manifest validity: 100% (8/8 valid, 0 errors, 0 warnings) · Surface score 70 (coverage 40 + breadth 15 + conformance 15). 9 standards observed across 4 of 9 steps; 8 manifests present (+ security.txt), 0 errors, 0 warnings. Checked 2026-10-25 23:11:47 UTC. This grades the agent-facing files published at belgravialabs.com, not the organisation's products.
Coverage by step
Manifests
| Manifest | Result | Findings |
|---|---|---|
Agentic Resource Discovery/.well-known/ard.json | valid | |
AI Catalog well-known file/.well-known/ai-catalog.json | valid | |
MCP Server Cards (SEP-2127)/.well-known/mcp/server-card.json | valid | |
Agent2Agent Protocol/.well-known/agent-card.json | valid | |
llms.txt/llms.txt | valid | |
AGENTS.md/AGENTS.md | valid | |
Agent Skills/.well-known/agent-skills/index.json | absent HTTP 404 | |
api-catalog: A Well-Known URI and Link Relation to Help Discovery of APIs/.well-known/api-catalog | valid | |
AI Crawler User-Agent Tokens in robots.txt/robots.txt | valid | |
security.txt (RFC 9116)/.well-known/security.txt | valid | |
auth.md/auth.md | absent HTTP 404 | |
Open Knowledge Format/okf/index.md | absent HTTP 404 |
What an agent can call
- MCP endpoint
https://agenticweb.wiki/mcp(streamable-http) — from/.well-known/mcp/server-card.json - A2A agent Belgravia Labs — AgenticWeb Verifier at
https://agenticweb.wiki/mcp, skills: Validate Domain Agentic Readiness, Lookup Agentic Web Standards - API description
https://belgravialabs.com/.well-known/mcp/server-card.json(application/mcp-server-card+json) - API description
https://agenticweb.wiki/connect/(text/html) - API description
https://agenticweb.wiki/openapi.json(application/vnd.oai.openapi+json;version=3.1) - API description
https://agenticweb.wiki/data/(text/html) - llms.txt: Belgravia Labs, 11 links
- robots.txt AI policy: allows GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, PerplexityBot; unlisted crawlers allowed; Agentmap https://belgravialabs.com/.well-known/ard.json
- ARD bridge (
/.well-known/ard.json): 2/3 valid, 0 legacy, 0 dead/broken
Standards observed
- Agent2Agent Protocol https://belgravialabs.com/.well-known/agent-card.json
- llms.txt https://belgravialabs.com/llms.txt
- AGENTS.md https://belgravialabs.com/AGENTS.md
- Content Signals Policy https://belgravialabs.com/robots.txt
- Agentic Resource Discovery https://belgravialabs.com/.well-known/ard.json
- AI Catalog well-known file https://belgravialabs.com/.well-known/ai-catalog.json
- api-catalog: A Well-Known URI and Link Relation to Help Discovery of APIs https://belgravialabs.com/.well-known/api-catalog
- AI Crawler User-Agent Tokens in robots.txt https://belgravialabs.com/robots.txt
- MCP Server Cards (SEP-2127)
Cheapest next steps
No cheap file is missing, or the origin never answered.
Agent pre-flight & practical readiness
Preferred mode: mcp (fallback a2a). Call the MCP endpoint at https://agenticweb.wiki/mcp first. Do not scrape the HTML when a tool can answer.
Fetch: POST https://agenticweb.wiki/mcp POST a JSON-RPC tools/list, or server/discover on the 2026-07-28 revision, before reading HTML.
Practical readability: exa_parsable yes, markdown negotiation no, render ssr_rich, HTML quality high, first content at byte 99848, HTML-to-llms.txt token ratio 60.42×, JSON-LD Organization.
AI crawler posture: search_allowed_training_blocked · Assistant access: OpenAI allowed, Anthropic allowed, Google allowed, Perplexity allowed, Exa allowed.
Readiness subscores: reach 20/20, readability 22/30, capabilities 25/35, conformance 15/15, total 82/100, grade A+.
How the grade is made
+10 per lifecycle step with a standard observed, counting manifests that validate or that the probe rule for the standard recognises (a valid MCP server card counts for MCP; max 90); +3 per observed standard beyond the first two (max 15); conformance +15 (+10 with one manifest) minus 5 per agent-facing manifest with errors, never below 0 (security.txt is reported, not scored). Lifecycle grade: A+ >= 60, A >= 45, B >= 30, C >= 20, D >= 10. Strict 100-point composite (subscores.total_100 / subscores.grade_100): reach_and_policy (20) + readability (30) + capabilities (35) + conformance (15), where A+ >= 80, A >= 65, B >= 45, C >= 28, D >= 15. Errors are specification violations; warnings are best-practice gaps. Only public paths are fetched, without credentials, so standards that need a login (skipped in the probe) are not counted either way. The same report is the validate_manifest tool on the MCP server.
JSON · Markdown · probe rules