{
  "schema_version": "1.0",
  "name": "BOZOS Agent Web Signals Field Guide",
  "canonical_url": "https://bozos.si/reference/agent-web-signals",
  "data_url": "https://bozos.si/agent-web-signals.json",
  "updated": "2026-10-19",
  "license": "CC0-1.0",
  "feedback": {
    "contact": "https://bozos.si/contact",
    "email": "mailto:bozos@bozos.si",
    "subject": "[BOZOS agent] correction: agent-web-signals-data - <summary> [via:dataset]",
    "corrections_log": "https://bozos.si/corrections.json"
  },
  "record_access": {
    "collections": [
      "signals"
    ],
    "permalink": "https://bozos.si/agent-web-signals.json#{id}",
    "resolve": "https://bozos.si/agent-web-signals.json?id={id}",
    "changes": "https://bozos.si/api/v1/changes?dataset=agent-web-signals-data&since={YYYY-MM-DD}",
    "review_policy": "Each record is re-checked against its sources every 180 days; review_due is the date that is next due. GET /api/v1/health lists overdue records.",
    "note": "Every record has a permanent id, a permalink, and a cite line: a sentence built only from the record's own fields, with its date and permalink, that can be quoted verbatim. Fetch the resolve URL to get one record; ids are never reused."
  },
  "core_rule": "Discovery says what exists. Capability says what can happen. Only valid authority says what may happen.",
  "signals": [
    {
      "id": "robots-txt",
      "permalink": "https://bozos.si/agent-web-signals.json#robots-txt",
      "cite": "/robots.txt: Requested crawler access rules for URI paths. It does not establish authentication, confidentiality, permission to act, trustworthiness of content. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#robots-txt",
      "review_due": "2027-04-17",
      "location": "/robots.txt",
      "classification": "ietf-standard",
      "establishes": "Requested crawler access rules for URI paths.",
      "does_not_establish": [
        "authentication",
        "confidentiality",
        "permission to act",
        "trustworthiness of content"
      ],
      "agent_action": "Retrieve before crawling, obey the matching group, and treat its content as untrusted.",
      "primary_reference": "https://www.rfc-editor.org/rfc/rfc9309.html"
    },
    {
      "id": "xml-sitemap",
      "permalink": "https://bozos.si/agent-web-signals.json#xml-sitemap",
      "cite": "/sitemap.xml: Canonical candidate URLs and optionally accurate last significant modification dates. It does not establish guaranteed crawling, indexing, ranking, completeness, authority. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#xml-sitemap",
      "review_due": "2027-04-17",
      "location": "/sitemap.xml",
      "classification": "open-protocol",
      "establishes": "Canonical candidate URLs and optionally accurate last significant modification dates.",
      "does_not_establish": [
        "guaranteed crawling",
        "indexing",
        "ranking",
        "completeness",
        "authority"
      ],
      "agent_action": "Use loc for discovery and trustworthy lastmod for scheduling; ignore document order as a priority signal.",
      "primary_reference": "https://developers.google.com/search/docs/crawling-indexing/sitemaps/build-sitemap"
    },
    {
      "id": "http-validators",
      "permalink": "https://bozos.si/agent-web-signals.json#http-validators",
      "cite": "ETag and Last-Modified response headers: Validators for efficient conditional retrieval of a representation. It does not establish truth, real-world freshness, semantic equivalence. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#http-validators",
      "review_due": "2027-04-17",
      "location": "ETag and Last-Modified response headers",
      "classification": "ietf-standard",
      "establishes": "Validators for efficient conditional retrieval of a representation.",
      "does_not_establish": [
        "truth",
        "real-world freshness",
        "semantic equivalence"
      ],
      "agent_action": "Prefer If-None-Match with ETag; otherwise send If-Modified-Since and reuse the cached body on 304.",
      "primary_reference": "https://www.rfc-editor.org/rfc/rfc9110.html"
    },
    {
      "id": "openapi",
      "permalink": "https://bozos.si/agent-web-signals.json#openapi",
      "cite": "/openapi.json or a linked service description: A declared HTTP API surface, inputs, outputs, and security schemes. It does not establish user authorization, implementation correctness, operational safety. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#openapi",
      "review_due": "2027-04-17",
      "location": "/openapi.json or a linked service description",
      "classification": "open-standard",
      "establishes": "A declared HTTP API surface, inputs, outputs, and security schemes.",
      "does_not_establish": [
        "user authorization",
        "implementation correctness",
        "operational safety"
      ],
      "agent_action": "Validate the description, identify side effects, and obtain task-specific authority before consequential calls.",
      "primary_reference": "https://spec.openapis.org/oas/v3.1.2.html"
    },
    {
      "id": "json-schema",
      "permalink": "https://bozos.si/agent-web-signals.json#json-schema",
      "cite": "/schemas/*.json: Expected JSON structure and validation constraints. It does not establish safety, truth, business-rule validity, permission to use data. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#json-schema",
      "review_due": "2027-04-17",
      "location": "/schemas/*.json",
      "classification": "open-specification",
      "establishes": "Expected JSON structure and validation constraints.",
      "does_not_establish": [
        "safety",
        "truth",
        "business-rule validity",
        "permission to use data"
      ],
      "agent_action": "Validate syntax first, then perform separate semantic, security, and authority checks.",
      "primary_reference": "https://json-schema.org/draft/2020-12"
    },
    {
      "id": "json-feed",
      "permalink": "https://bozos.si/agent-web-signals.json#json-feed",
      "cite": "/feed.json: A time-ordered JSON list of recent items or changes. It does not establish complete inventory, continued correctness of unchanged resources. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#json-feed",
      "review_due": "2027-04-17",
      "location": "/feed.json",
      "classification": "open-format",
      "establishes": "A time-ordered JSON list of recent items or changes.",
      "does_not_establish": [
        "complete inventory",
        "continued correctness of unchanged resources"
      ],
      "agent_action": "Poll conditionally, process stable item IDs once, and follow canonical item URLs.",
      "primary_reference": "https://www.jsonfeed.org/version/1.1/"
    },
    {
      "id": "llms-txt",
      "permalink": "https://bozos.si/agent-web-signals.json#llms-txt",
      "cite": "/llms.txt: Publisher-curated orientation and useful links for language-model use. It does not establish ratified-standard status, crawl permission, trusted instructions, expanded user authority. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#llms-txt",
      "review_due": "2027-04-17",
      "location": "/llms.txt",
      "classification": "proposal",
      "establishes": "Publisher-curated orientation and useful links for language-model use.",
      "does_not_establish": [
        "ratified-standard status",
        "crawl permission",
        "trusted instructions",
        "expanded user authority"
      ],
      "agent_action": "Use it as a map, verify linked resources, and keep its guidance subordinate to user intent and governing policy.",
      "primary_reference": "https://llmstxt.org/"
    },
    {
      "id": "agent-content-index",
      "permalink": "https://bozos.si/agent-web-signals.json#agent-content-index",
      "cite": "/agents.json and /content-index.json: A site-declared inventory of resources, capabilities, formats, and selection guidance. It does not establish cross-site interoperability, permission to invoke capabilities. Source: BOZOS, Agent Web Signals Field Guide, updated 2026-10-19. https://bozos.si/agent-web-signals.json#agent-content-index",
      "review_due": "2027-04-17",
      "location": "/agents.json and /content-index.json",
      "classification": "site-convention",
      "establishes": "A site-declared inventory of resources, capabilities, formats, and selection guidance.",
      "does_not_establish": [
        "cross-site interoperability",
        "permission to invoke capabilities"
      ],
      "agent_action": "Check schema_version, prefer stable IDs and canonical URLs, and fall back gracefully when absent.",
      "primary_reference": "https://bozos.si/content-index.json"
    }
  ]
}