REST API / Collections

`considered_sources`

CollectionLarge: filter on indexed fields

Uncited source candidates associated with an AI answer. ENGINE_GROUNDED is engine-observed grounding; SERP_ORGANIC and RELATED_QUESTION are observed search candidates, not proof the AI read a page. SYNTHETIC_SERP is synthetic search, never AI retrieval. Filter discovered_via before computing rates. Outside text follows the mentions masking rules.

Max rows per query: 100
Default order: id asc
Required filter: one of project_id, id, answer_id
Max offset: 2000
Cursor paging: by id

Use read report_* for dashboard headline metrics and anchored comparison windows.

Fields

Field	Type	Description	Filter	Sort
`id` indexed	`string`	Unique row ID.	✓	✓
`project_id` indexed	`string`	Project ID from projects.	✓	–
`answer_id` indexed	`string`	AI answer ID.	✓	✓
`url` nullable redacted	`string`	Source URL, or null when it identifies a personal account.	–	–
`domain` indexed	`string`	Source domain.	✓	✓
`title` nullable redacted	`string`	Source title; null for LinkedIn sources.	–	–
`snippet` nullable redacted	`string`	Source excerpt; null for LinkedIn sources.	–	–
`discovered_via` indexed	`string`	How the source was discovered. `ENGINE_GROUNDED` `SERP_ORGANIC` `RELATED_QUESTION` `SYNTHETIC_SERP`	✓	✓
`serp_query` nullable	`string`	Search query that surfaced the source.	–	–
`serp_position` nullable	`integer`	Search result position.	–	–
`category` nullable	`string`	Source category. `EDITORIAL` `REVIEW` `DIRECTORY` `BRAND` `WIKIPEDIA` `SOCIAL` `COMMUNITY` `UGC` `OTHER`	–	–
`page_type` nullable indexed	`string`	Page type.	✓	✓
`published_at` nullable	`datetime`	Source publication time.	–	–
`redacted_fields`	`json`	Populated fields hidden by privacy policy, with a reason. Empty means no populated field was redacted.	–	–

Examples

Example

SELECT id, project_id, answer_id, url, domain
FROM considered_sources
WHERE project_id = $1
ORDER BY id ASC
LIMIT 25

Calling it

MCP

Tool call

query({"sql":"SELECT id, project_id, answer_id, url, domain FROM considered_sources WHERE project_id = $1 ORDER BY id ASC LIMIT 25","params":["<value 1>"]})

REST

curl

curl -X POST "https://app.sensorhub.ai/api/v1/query" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  --data-binary @- <<'JSON'
{
  "sql": "SELECT id, project_id, answer_id, url, domain FROM considered_sources WHERE project_id = $1 ORDER BY id ASC LIMIT 25",
  "params": [
    "<value 1>"
  ]
}
JSON

Row JSON Schema

json

{
  "type": "object",
  "description": "Uncited source candidates associated with an AI answer. ENGINE_GROUNDED is engine-observed grounding; SERP_ORGANIC and RELATED_QUESTION are observed search candidates, not proof the AI read a page. SYNTHETIC_SERP is synthetic search, never AI retrieval. Filter discovered_via before computing rates. Outside text follows the mentions masking rules.",
  "properties": {
    "id": {
      "type": "string",
      "description": "Unique row ID.",
      "x-sensorhub-indexed": true,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": true,
      "x-sensorhub-sortable": true
    },
    "project_id": {
      "type": "string",
      "description": "Project ID from projects.",
      "x-sensorhub-indexed": true,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": false,
      "x-sensorhub-filterable": true,
      "x-sensorhub-sortable": false
    },
    "answer_id": {
      "type": "string",
      "description": "AI answer ID.",
      "x-sensorhub-indexed": true,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": true,
      "x-sensorhub-sortable": true
    },
    "url": {
      "type": [
        "string",
        "null"
      ],
      "description": "Source URL, or null when it identifies a personal account.",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": false,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "domain": {
      "type": "string",
      "description": "Source domain.",
      "x-sensorhub-indexed": true,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": true,
      "x-sensorhub-sortable": true
    },
    "title": {
      "type": [
        "string",
        "null"
      ],
      "description": "Source title; null for LinkedIn sources.",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": false,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "snippet": {
      "type": [
        "string",
        "null"
      ],
      "description": "Source excerpt; null for LinkedIn sources.",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": false,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "discovered_via": {
      "type": "string",
      "description": "How the source was discovered.",
      "enum": [
        "ENGINE_GROUNDED",
        "SERP_ORGANIC",
        "RELATED_QUESTION",
        "SYNTHETIC_SERP"
      ],
      "x-sensorhub-filter-operators": [
        "=",
        "!=",
        "<>",
        "IN",
        "NOT IN",
        "IS NULL",
        "IS NOT NULL"
      ],
      "x-sensorhub-indexed": true,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": true,
      "x-sensorhub-sortable": true
    },
    "serp_query": {
      "type": [
        "string",
        "null"
      ],
      "description": "Search query that surfaced the source.",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "serp_position": {
      "type": [
        "integer",
        "null"
      ],
      "description": "Search result position.",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "category": {
      "type": [
        "string",
        "null"
      ],
      "description": "Source category.",
      "enum": [
        "EDITORIAL",
        "REVIEW",
        "DIRECTORY",
        "BRAND",
        "WIKIPEDIA",
        "SOCIAL",
        "COMMUNITY",
        "UGC",
        "OTHER",
        null
      ],
      "x-sensorhub-filter-operators": [],
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "page_type": {
      "type": [
        "string",
        "null"
      ],
      "description": "Page type.",
      "x-sensorhub-indexed": true,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": true,
      "x-sensorhub-sortable": true
    },
    "published_at": {
      "type": [
        "string",
        "null"
      ],
      "description": "Source publication time.",
      "format": "date-time",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": true,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    },
    "redacted_fields": {
      "description": "Populated fields hidden by privacy policy, with a reason. Empty means no populated field was redacted.",
      "x-sensorhub-indexed": false,
      "x-sensorhub-searchable": false,
      "x-sensorhub-aggregatable": false,
      "x-sensorhub-filterable": false,
      "x-sensorhub-sortable": false
    }
  },
  "additionalProperties": false
}