Unified semantic search across multiple entity types.

​

Overview

This endpoint provides semantic search across your accessible data. It accepts natural language queries and returns results ranked by relevance.

Each result contains:

  • type: entity type, e.g. "contact" or "company"
  • id: entity ID
  • item: entity payload (shape depends on type)

Additional entity types may be added over time. Clients should:

  • route rendering by type
  • ignore unknown types gracefully
  • use the optional types filter if strict type control is needed

Usage Examples

Search across all currently supported types:

{
  "query": "software engineers in Stockholm",
  "limit": 25
}

Search a strict subset of types:

{
  "query": "investors interested in logistics properties in Northern Europe",
  "types": ["contact", "company"],
  "limit": 10
}
Body·
required

Search request with natural language query and optional type filter

Request body for unified semantic search.

  • query
    Type: string
    min length:  
    1
    required

    Natural language query.

  • limit
    Type: numberFormat: int32
    min:  
    1
    max:  
    100

    Maximum number of results to return.

  • types
    Type: array string[] …10

    Optional result type filter.

Responses
  • application/json
  • application/problem+json
  • application/problem+json
  • application/problem+json
  • application/problem+json
Request Example for post/api/v1/search
curl https://app.propstreet.com/api/v1/search \
  --request POST \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Bearer YOUR_SECRET_TOKEN' \
  --data '{
  "query": "",
  "limit": 1,
  "types": [
    ""
  ]
}'

{
  "results": [
    {
      "type": "string",
      "id": "string",
      "score": "string",
      "item": null
    }
  ],
  "total": "string"
}

Returns ranked search results

Propstreet API Documentation