val.town agentic web readiness

C Manifest validity: 50% (1/2 valid, 1 error, 5 warnings) · Surface score 20 (coverage 10 + breadth 0 + conformance 10). 2 standards observed across 1 of 9 steps; 2 manifests present (+ security.txt), 1 error, 5 warnings. Checked 2026-10-25 13:54:11 UTC via www.val.town. This grades the agent-facing files published at val.town, not the organisation's products.

Badge for a README: ![agentic web grade](https://agenticweb.wiki/badge/val.town.svg) · JSON · Markdown · Re-check now

Coverage by step

Manifests

# search and answer-engine crawlers
User-agent: Googlebot
User-agent: bingbot
User-agent: OAI-SearchBot
User-agent: Claude-SearchBot
User-agent: PerplexityBot
User-agent: ExaSearchBot
User-agent: TavilyBot
Allow: /

# model-training crawlers: choose Allow or Disallow and keep it explicit
User-agent: GPTBot
User-agent: ClaudeBot
User-agent: Google-Extended
User-agent: Applebot-Extended
User-agent: CCBot
User-agent: meta-externalagent
User-agent: Bytespider
Disallow: /

# user-triggered fetchers (a person asked for this page)
User-agent: ChatGPT-User
User-agent: Claude-User
User-agent: Perplexity-User
Allow: /

# default fallback for unlisted crawlers (RFC 9309 §2.2.1: specific groups above take precedence over *)
User-agent: *
Allow: /
Disallow: /account/
Content-Signal: ai-train=no, search=yes, ai-input=yes

Sitemap: https://val.town/sitemap.xml
Agentmap: https://val.town/.well-known/ard.json

What an agent can call

• llms.txt: Val Town, 5 links
• robots.txt AI policy: no AI crawler tokens named; unlisted crawlers unspecified

Standards observed

• llms.txt https://val.town/llms.txt
• Markdown for Agents https://val.town/

Cheapest next steps

1. Agentic Resource Discovery: a catalog at /.well-known/ard.json lets agents discover your APIs, MCP servers and data in one fetch (emerging spec — publish only if you run the underlying endpoints; generate via MCP template({"id": "ard", "domain": "val.town"}))
2. AGENTS.md: instructions for coding agents working on or with the site
   Generate starter /AGENTS.md for val.town

   # Val for agents
   
   Val and what it offers to agents
   
   ## Read the data, not the HTML
   - Index: https://val.town/llms.txt
   - API description: https://val.town/openapi.json
   - Discovery catalog: https://val.town/.well-known/ard.json
   
   ## Calling
   - MCP server (Streamable HTTP): https://val.town/mcp
   - Rate limits and authentication: describe them here
   
   ## Conventions
   - How to cite or attribute content
   - What not to do (no scraping of /account/, no automated sign-ups, etc.)
   
   Contact: security@val.town
   

3. api-catalog: A Well-Known URI and Link Relation to Help Discovery of APIs: an RFC 9727 linkset pointing at your OpenAPI description (conditional spec — publish only if you run the underlying endpoints; generate via MCP template({"id": "api-catalog-rfc9727", "domain": "val.town"}))

Agent pre-flight & practical readiness

Preferred mode: llms_txt (fallback markdown_negotiation). GET /llms.txt and follow the links it names. That index is the page list an agent should read. Split-brain llms.txt: val.town/llms.txt has 5 links while docs.val.town/llms.txt has 119 technical links. Probe docs.val.town for full developer documentation.

Fetch: GET https://val.town/llms.txt GET /llms.txt and follow its links.

Subdomain discovery hint (split_brain_llms_txt): Split-brain llms.txt: val.town/llms.txt has 5 links while docs.val.town/llms.txt has 119 technical links. Probe docs.val.town for full developer documentation. (check docs.val.town)

Practical readability: exa_parsable yes, markdown negotiation yes, render markdown_native, HTML quality high, first content at byte 0, HTML-to-llms.txt token ratio 6.42×.

AI crawler posture: unspecified · Assistant access: OpenAI unspecified, Anthropic unspecified, Google unspecified, Perplexity unspecified, Exa unspecified.

Readiness subscores: reach 2/20, readability 30/30, capabilities 6/35, conformance 10/15, total 48/100, grade B.

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