AI Discovery Files

Tureka AI Visibility Checker uses this page to explain how this signal affects AI visibility reports and which implementation details matter most.

llms.txt can summarize canonical pages, services, policies, and contact details for AI systems. Tureka also checks related discovery files: llm.txt, llms.html, ai.txt, ai.json, identity.json, brand.txt, faq-ai.txt, developer-ai.txt, and robots-ai.txt.

These alternate files are emerging conventions. Missing files are treated as low-impact GEO warnings, while llms.txt remains the strongest general-purpose signal.

faq-ai.txt should contain answer-ready questions and factual answers. developer-ai.txt can describe API usage, implementation limits, and integration notes. robots-ai.txt can separate AI-specific crawler guidance from general robots.txt rules.

Agent-ready sites can also expose /.well-known/api-catalog, /.well-known/mcp/server-card.json, /.well-known/agent-skills/index.json, OAuth discovery metadata, OAuth protected resource metadata, and /auth.md when APIs or agent actions exist.

The Tureka API catalog now advertises a free web scan plus two x402 agent packages: agent-readiness-validator at $0.15 and visibility-score at $0.05. Agents can discover these through OpenAPI, llms.txt, API catalog, MCP server card, and Agent Skills metadata.

Homepage Link headers can advertise discovery resources without changing visible page content. Markdown negotiation can give agents a cleaner text version while browsers keep receiving HTML.

Do not use AI discovery files to invent claims. They should repeat factual, visible, canonical information from the website.

The report can now generate starter snippets for llms.txt, robots.txt AI crawler rules, and Organization JSON-LD from the scan's extracted signals.

identity.json and brand.txt remain optional, but if they exist the scanner checks basic parseability and useful minimum content.

Copyable snippets are starter files, not guarantees. Review them before publishing and keep them aligned with visible site content.

Criteria covered here

llms.txt is missingllms.txt is thin or missingllm.txt is missingllms.html is missingai.txt is missingai.json is missingidentity.json does not parse cleanlyidentity.json is missingbrand.txt is thinbrand.txt is missingfaq-ai.txt is missingdeveloper-ai.txt is missingrobots-ai.txt is missingAI Plugin JSON is missingOpenAPI Spec is missingAPI catalog is missingAPI catalog needs linkset fieldsMCP Discovery JSON is missingMCP Discovery JSON needs clearer fieldsMCP Server Card is missingMCP Server Card fields are incompleteAgent Skills index is missingAgent Skills entries are incompleteOAuth/OIDC discovery metadata is incompleteOAuth protected resource metadata is incompleteauth.md is missing