AI Intelligence
Verification

Agentic verification search

Search returning frozen verification schema results. Designed for AI agents, ATS systems, and background check integrations. Requires API key (not JWT). SCRUM-3906: this route now answers with EITHER a semantic (embedding + cosine-similarity) match OR a lexical (ILIKE substring) match — search_mode on the response says which. Semantic is used when ENABLE_SEMANTIC_SEARCH is on and the embed+match RPC succeeds; the route falls back to lexical, never a 503, when the flag is off or that RPC/embedding call fails. Only the semantic path costs an AI credit or emits similarity; issuer_name, issued_date, expiry_date and anchor_timestamp are omitted (never null) on a lexical result — that path does not query them. Semantic results always include anchor_timestamp: the Network Observed Time, or null when no network observation is available; creation time is never substituted.

get/verify/search

Request

  • Base URL: https://app.arkova.ai/api/v1
  • URL: https://app.arkova.ai/api/v1/verify/search
  • Auth: one of:
    • HTTP bearer
    • API key in header X-API-Key

Query parameters

qstring required
thresholdnumber

Semantic-path only; ignored on a lexical-path request.

limitinteger

Response

Verification results. search_mode (semantic_vector | lexical_substring) tells the caller which path answered — see the operation description.

querystring
countinteger
thresholdnumber

Echoes the request threshold; not applied on the lexical path.

search_mode'semantic_vector' | 'lexical_substring'

Example response

{
  "results": [
    {
      "compliance_controls": [
        "SOC2-CC6.1",
        "GDPR-5.1f",
        "FERPA-99.31"
      ]
    }
  ]
}

Changes

No recorded changes to this operation across all 1 revision of this API.