tavily.com agentic web readiness

A+ Manifest validity: 100% (7/7 valid, 0 errors, 10 warnings) · Surface score 70 (coverage 40 + breadth 15 + conformance 15). 7 standards observed across 4 of 9 steps; 7 manifests present, 0 errors, 10 warnings. Checked 2026-10-25 21:16:58 UTC via www.tavily.com. This grades the agent-facing files published at tavily.com, not the organisation's products.

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

Coverage by step

Manifests

What an agent can call

• MCP endpoint https://mcp.tavily.com/mcp/ (streamable-http) — from /.well-known/mcp.json
• A2A agent Tavily at https://api.tavily.com, skills: Web Search, Content Extract, Web Crawl, Site Map, Deep Research
• A2A agent Tavily at https://api.tavily.com, skills: Web Search, Content Extract, Web Crawl, Site Map, Deep Research
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://docs.tavily.com/documentation/api-reference/introduction (text/html)
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://docs.tavily.com/documentation/api-reference/endpoint/search (text/html)
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://docs.tavily.com/documentation/api-reference/endpoint/extract (text/html)
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://docs.tavily.com/documentation/api-reference/endpoint/crawl (text/html)
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://docs.tavily.com/documentation/api-reference/endpoint/map (text/html)
• API description https://tavily.com/.well-known/openapi.json (application/json)
• API description https://docs.tavily.com/documentation/api-reference/endpoint/research (text/html)
• API description https://tavily.com/.well-known/mcp.json (application/json)
• API description https://docs.tavily.com/documentation/mcp (text/html)
• llms.txt: Tavily Docs, 118 links, llms-full.txt present
• robots.txt AI policy: unlisted crawlers partial
• Endpoint check (mcp https://mcp.tavily.com/mcp/): auth_required (HTTP 401 (authenticated MCP endpoint: Bearer scope="openid offline_access", resource_metadata="https://mcp.tavily.com/))
• Endpoint check (a2a https://api.tavily.com): dead (HTTP 405)
• ARD bridge (/.well-known/ard.json): 0/1 valid, 0 legacy, 0 dead/broken

Standards observed

• llms.txt https://tavily.com/llms.txt
• AGENTS.md https://tavily.com/AGENTS.md
• Agentic Resource Discovery https://tavily.com/.well-known/ard.json
• api-catalog: A Well-Known URI and Link Relation to Help Discovery of APIs https://tavily.com/.well-known/api-catalog
• AI Crawler User-Agent Tokens in robots.txt https://tavily.com/robots.txt
• MCP Server Cards (SEP-2127)
• Model Context Protocol

Cheapest next steps

1. security.txt (RFC 9116): RFC 9116 contact and expiry, read by agents and people
   Generate starter /.well-known/security.txt for tavily.com

   Contact: mailto:security@tavily.com
   Expires: 2027-09-20T23:44:21Z
   Canonical: https://tavily.com/.well-known/security.txt
   Preferred-Languages: en
   Policy: https://tavily.com/security-policy
   

Agent pre-flight & practical readiness

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

Fetch: POST https://mcp.tavily.com/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 36565, HTML-to-llms.txt token ratio 8.348×, JSON-LD Organization, ImageObject, Place, Person, ContactPoint, PostalAddress, WebSite, Service, Audience, Offer, Product, Brand, SoftwareApplication, WebAPI, WebPage.

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

Readiness subscores: reach 20/20, readability 22/30, capabilities 27/35, conformance 15/15, total 84/100, 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