openrouter.ai agentic web readiness

A Manifest validity: 75% (3/4 valid, 1 error, 9 warnings) · Surface score 52 (coverage 30 + breadth 12 + conformance 10). 6 standards observed across 3 of 9 steps; 4 manifests present (+ robots.txt, security.txt), 1 error, 9 warnings. Checked 2026-10-25 08:50:11 UTC. This grades the agent-facing files published at openrouter.ai, not the organisation's products.

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

Coverage by step

Manifests

{
  "specVersion": "1.0",
  "host": {
    "displayName": "Openrouter",
    "identifier": "did:web:openrouter.ai",
    "documentationUrl": "https://openrouter.ai/AGENTS.md"
  },
  "entries": [
    {
      "identifier": "urn:air:openrouter.ai:mcp:openrouter",
      "displayName": "Openrouter MCP server",
      "type": "application/mcp-server-card+json",
      "url": "https://openrouter.ai/mcp/server-card",
      "description": "Openrouter and what it offers to agents. Streamable HTTP MCP server.",
      "representativeQueries": [
        "what can the Openrouter MCP server do",
        "use Openrouter tools from an agent"
      ],
      "tags": [
        "mcp",
        "mcp-server"
      ]
    },
    {
      "identifier": "urn:air:openrouter.ai:api:openapi",
      "displayName": "Openrouter API description",
      "type": "application/openapi+json",
      "url": "https://openrouter.ai/openapi.json",
      "description": "OpenAPI description of the Openrouter API.",
      "representativeQueries": [
        "Openrouter API reference"
      ],
      "tags": [
        "openapi",
        "api"
      ]
    }
  ]
}

What an agent can call

• MCP endpoint https://mcp.openrouter.ai/mcp (streamable-http), tools: ping, list-models, get-model, list-model-endpoints, list-providers, get-endpoint-uptime-history, list-benchmarks, list-task-classifications, list-daily-model-rankings, list-app-rankings, list-presets, get-preset, get-credits, get-generation, send-feedback, search-docs, send-message, generate-image, transcribe-audio, generate-speech — from /.well-known/mcp/server-card.json
• API description https://openrouter.ai/openapi.json (application/vnd.oai.openapi+json)
• API description https://openrouter.ai/openapi.json (service-desc)
• API description https://openrouter.ai/docs/api_reference/overview (service-doc)
• llms.txt: OpenRouter, 42 links, llms-full.txt present
• robots.txt AI policy: no AI crawler tokens named; unlisted crawlers allowed on public paths (subpaths disallowed under User-agent: *)
• ARD bridge (/.well-known/ard.json): 1/3 valid, 0 legacy, 0 dead/broken

Standards observed

• llms.txt https://openrouter.ai/llms.txt
• Agentic Resource Discovery https://openrouter.ai/.well-known/ard.json
• api-catalog: A Well-Known URI and Link Relation to Help Discovery of APIs https://openrouter.ai/.well-known/api-catalog
• Markdown for Agents https://openrouter.ai/
• MCP Server Cards (SEP-2127)
• Model Context Protocol

Cheapest next steps

1. AGENTS.md: instructions for coding agents working on or with the site
   Generate starter /AGENTS.md for openrouter.ai

   # Openrouter for agents
   
   Openrouter and what it offers to agents
   
   ## Read the data, not the HTML
   - Index: https://openrouter.ai/llms.txt
   - API description: https://openrouter.ai/openapi.json
   - Discovery catalog: https://openrouter.ai/.well-known/ard.json
   
   ## Calling
   - MCP server (Streamable HTTP): https://openrouter.ai/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@openrouter.ai
   

Agent pre-flight & practical readiness

Preferred mode: mcp (fallback openapi). Call the MCP endpoint at https://mcp.openrouter.ai/mcp first. Do not scrape the HTML when a tool can answer.

Fetch: POST https://mcp.openrouter.ai/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 yes, render markdown_native, HTML quality high, first content at byte 0.

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

Readiness subscores: reach 19/20, readability 30/30, capabilities 26/35, conformance 10/15, total 85/100, 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