generated: '2026-08-05' method: searched source: https://veir.com/sitemap.md + https://veir.com/.well-known/api-catalog + https://veir.com/robots.txt summary: >- VEIR ships no product API. What it does ship is a deliberate agent-facing content layer: a markdown twin of every marketing page, a semantic index at /sitemap.md, an RFC 9727 API catalog pointing at it, and a machine-readable AI usage preference in robots.txt. These are the only cross-cutting conventions the public surface exposes. authentication: style: none-required note: >- Every documented surface below is anonymous. An OIDC discovery document exists but its endpoints do not resolve — see authentication/veir-authentication.yml. markdown_twins: supported: true convention: append `.md` to a page path to receive that page's full content as text/markdown index: https://veir.com/sitemap.md index_declares: >- "This file is a semantic index of veir.com. It is intended for LLM-assisted navigation and agent discovery. Each entry links to a `.md` endpoint that returns the full page content as markdown." index_entries: 39 content_type: text/markdown; charset=utf-8 verified: - url: https://veir.com/about.md status: 200 content_type: text/markdown bytes: 16673 - url: https://veir.com/news.md status: 200 content_type: text/markdown coverage_note: >- Only paths listed in /sitemap.md have twins. /technology.md returned 404 ("Page not found") even though https://veir.com/technology/ is a live page, so the twin set is the sitemap set, not the whole site. api_catalog: standard: RFC 9727 url: https://veir.com/.well-known/api-catalog anchor: https://veir.com/api/markdown service_desc: https://veir.com/sitemap.md service_doc: https://veir.com/sitemap.md defects: - >- The catalog's service-desc and service-doc both point at /sitemap.md, a human/LLM index in markdown. RFC 9727 expects service-desc to be a machine-readable API description (OpenAPI, AsyncAPI); a markdown sitemap does not describe an interface. - >- The anchored endpoint https://veir.com/api/markdown returns 200 text/markdown but ignores its input: /api/markdown?path=/about and /api/markdown?url=/about both returned the byte-identical homepage document (4394 bytes) as the bare /api/markdown. Verified 2026-08-05. - >- robots.txt contains "Disallow: /api/", which disallows the very endpoint the site's own api-catalog advertises as its service anchor. ai_usage_preference: source: https://veir.com/robots.txt directive: 'Content-Signal: ai-train=no, search=yes, ai-input=no' parsed: ai-train: false search: true ai-input: false tension: >- ai-input=no asks that page content not be used as input to generative AI responses, while /sitemap.md states the markdown endpoints exist "for LLM-assisted navigation and agent discovery". The two signals point in opposite directions; recorded as published, not resolved. pagination: supported: false idempotency: supported: false note: No write surface exists, so there is no idempotency contract to record. versioning: scheme: none-published rate_limiting: documented: false signalled_headers: [] error_envelope: format: html note: Unknown paths return a Next.js HTML 404; /technology.md returned a text/plain "Page not found". request_tracing: header: null related: - well-known/veir-well-known.yml - authentication/veir-authentication.yml - conformance/veir-conformance.yml - security/veir-domain-security.yml x-evidence: fetched: '2026-08-05' probes: - url: https://veir.com/sitemap.md status: 200 - url: https://veir.com/api/markdown status: 200 - url: https://veir.com/about.md status: 200 - url: https://veir.com/technology.md status: 404 - url: https://veir.com/robots.txt status: 200