Collections

Collections your key can see. Collections are created in the app.

Base URL https://futuresmart.ai/document-intelligence/api · every request sends x-api-key. Samples read the key from DOCUMENT_INTELLIGENCE_API_KEY. All endpoints

List your collections

GET/v1

List knowledge services. If owner_id is provided, returns every collection that account can reach — owned and shared alike — as bare ids. If not provided, returns all non-deleted knowledge services (master API key only).

include=summary exists because a Home-style "one card per collection" UI otherwise has to call GET /config and GET /enrichment per collection just to render a card — 1 + 2N requests for N collections, with no document count available at all no matter how many extra calls are made (no endpoint returns one). This aggregates that in one extra pass over Document/FieldConfiguration/KnowledgeConfig, grouped by knowledge_id, so it's a fixed number of queries regardless of N. See docs/screen-api-flow-document-intelligence.md §3 (G2+G3).

Includes collection_name (G1) — null for a collection created before that column existed, or created without one.

Owned vs. shared is deliberately NOT distinguished in this response (no owner_id per item either — see docs/api/knowledge-invitations.md). The UI's own Share flow calls GET /v1/validate ?owner_id=…&knowledge_id=… once per id from this list to split them into its "My collections" / "Shared with me" tabs — that is the intended discovery contract this endpoint feeds, not a separate "shared" listing.

Credential Document Intelligence API key (x-api-key)

Query parameters

Name	Type	Description
`owner_id`	`string \| null`
`include`	`string \| null`	Set to 'summary' to get per-collection document count, processing rollup, and field counts inline instead of a bare id list. Omit to keep the original bare-id response.

Other responses

422 Validation Error

import os
import requests

API_ROOT = "https://futuresmart.ai/document-intelligence/api"
API_KEY = os.environ["DOCUMENT_INTELLIGENCE_API_KEY"]

r = requests.get(
    f"{API_ROOT}/v1",
    headers={"x-api-key": API_KEY},
    timeout=120,
)
r.raise_for_status()
print(r.json())

const API_ROOT = "https://futuresmart.ai/document-intelligence/api";

const res = await fetch(`${API_ROOT}/v1`, {
  method: "GET",
  headers: { "x-api-key": process.env.DOCUMENT_INTELLIGENCE_API_KEY },
});
if (!res.ok) throw new Error(`${res.status} ${await res.text()}`);
console.log(await res.json());

curl -X GET "https://futuresmart.ai/document-intelligence/api/v1" \
  -H "x-api-key: $DOCUMENT_INTELLIGENCE_API_KEY"

Get a collection's configuration

GET/v1/{knowledge_id}/config

Credential Document Intelligence API key (x-api-key)

Path parameters

Name	Type	Description
`knowledge_id`required	`string`

Response · 200

Name	Type	Description
`knowledge_id`required	`string`
`embedding_model`	`string \| null`	One of: `paraphrase-multilingual-mpnet-base-v2`, `openai-text-embedding-3-large`, `openai-text-embedding-3-small`
`hybrid_search_enabled`	`boolean \| null`
`chunking_strategy`	`object \| null`	Chunking strategy configuration for knowledge settings.
`↳ document_level_metadata`	`boolean \| null`	Enable document-level metadata extractionDefault: true
`↳ chunk_level_metadata`	`boolean \| null`	Enable chunk-level metadata extractionDefault: false
`collection_name`	`string \| null`	Short display name (G1) — distinct from service_description, the longer free-text description. Null if never set.
`service_description`	`string \| null`
`enrichment_provider`	`string \| null`
`enrichment_model`	`string \| null`
`enrichment_temperature`	`number \| null`
`enrichment_prompt`	`string \| null`
`enrichment_source`	`string \| null`
`enrichment_enabled`	`boolean \| null`	Default: true
`use_deterministic_chunking`	`boolean \| null`	Default: true
`created_at`	`string \| null`
`updated_at`	`string \| null`

Other responses

422 Validation Error

import os
import requests

API_ROOT = "https://futuresmart.ai/document-intelligence/api"
API_KEY = os.environ["DOCUMENT_INTELLIGENCE_API_KEY"]

r = requests.get(
    f"{API_ROOT}/v1/YOUR_COLLECTION_ID/config",
    headers={"x-api-key": API_KEY},
    timeout=120,
)
r.raise_for_status()
print(r.json())

const API_ROOT = "https://futuresmart.ai/document-intelligence/api";

const res = await fetch(`${API_ROOT}/v1/YOUR_COLLECTION_ID/config`, {
  method: "GET",
  headers: { "x-api-key": process.env.DOCUMENT_INTELLIGENCE_API_KEY },
});
if (!res.ok) throw new Error(`${res.status} ${await res.text()}`);
console.log(await res.json());

curl -X GET "https://futuresmart.ai/document-intelligence/api/v1/YOUR_COLLECTION_ID/config" \
  -H "x-api-key: $DOCUMENT_INTELLIGENCE_API_KEY"

Response shape (200)

JSON

{
  "chunking_strategy": {
    "chunk_level_metadata": false,
    "document_level_metadata": true
  },
  "created_at": "2025-09-30T17:55:00Z",
  "embedding_model": "paraphrase-multilingual-mpnet-base-v2",
  "enrichment_model": "gpt-4.1-mini",
  "enrichment_prompt": "Extract metadata deterministically. Use only information explicitly present in the document. Do not infer missing facts. Return stable, normalized values that follow the field definitions exactly.",
  "enrichment_provider": "openai",
  "enrichment_temperature": 0,
  "hybrid_search_enabled": true,
  "knowledge_id": "123e4567-e89b-12d3-a456-426614174000",
  "service_description": "Knowledge base for customer support documents",
  "updated_at": "2025-09-30T17:55:00Z"
}