Skip to main content

​
Adding Data to Qontext

To retrieve useful context from Qontext, your first step is to add data to your Context Vault. You can do this in multiple ways: either automatically via native integrations or by using the Qontext API. Each method lets you bring your company knowledge into Qontext, making it immediately AI-ready for workflows, retrieval, and LLMs.

​
Ingest with Native Data Connections

Qontext connects directly to the tools your company already uses. By setting up these data connections, you can automatically add data from these sources to your Context Vault – no manual work required. Once connected, data is structured and linked, ready for retrieval through our API in your codebase or workflow builders.
  • HubSpot: Contacts, companies, deals, and interactions.
  • Google Drive: Documents, spreadsheets, presentations, and files.
  • Notion: Pages and databases
  • Gong: Meeting transcripts and call summaries.
  • Zendesk (coming soon): Support tickets, articles, and conversations.
  • Confluence (coming soon): Pages and team notes.
  • Intercom (coming soon): Conversations, leads, contacts, and messages
  • Pipedrive (coming soon): Deals, contacts, activities, and notes
If your tool isn’t listed, you can request it here. We’ll try our best to build a connection for you. In the meantime, you can still ingest data manually using our API (described below).

​
Manual Ingestion via API

If your data lives in a tool we don’t yet integrate with, or you want more control over which data flows into your context vaults, you can add data directly to Qontext using our API. This lets you:
  • Add Unstructured text: Add specific facts, notes, or scattered information directly to a Context Vault.
  • Add Unstructured text with metadata: Add specific facts with information about the provenance of your data.
  • Ingest Website content: Provide a URL to ingest one specific page (and, optionally, all linked pages excluding external links).
All ingestion requests require your workspaceId, knowledgeGraphId, and X-API-Key.
Use workflow builders such as n8n to automate ingestion from various sources into Context Vaults. See our n8n data ingestion guide.
Example: Ingest Unstructured Text 
curl --request POST \
  --url https://api.qontext.ai/v1/ingestion/unstructured \
  --header 'Content-Type: application/json' \
  --header 'X-API-Key: q_sk_***************' \
  --data '{
    "workspaceId": "ws_123e4567-e89b-12d3-a456-426614174000",
    "knowledgeGraphId": "kg_681c0068-b589-4923-ac4a-73040fab7a45",
    "text": "Our team will launch a new product on October 25, 2025."
  }'
Response: 
{
  "syncLogId": "sync_3f9b8e2a-1c8d-4e56-bbdf-123456789abc"
}
Example: Ingest Unstructured Text with Metadata 
curl --request POST \
  --url https://api.qontext.ai/v1/ingestion/any \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "workspaceId": "ws_123e4567-e89b-12d3-a456-426614174000",
    "knowledgeGraphId": "kg_681c0068-b589-4923-ac4a-73040fab7a45",
    "string_data": "Harry Potter created a new deal for Gryffindor Inc. in HubSpot.",
    "source": {
      "source_integration": "HubSpot",
      "source_data_type": "Deal",
      "source_id": "4754016823"
    }
  }'
Response: 
{
  "syncLogId": "sync_b5c7f1d2-3a4b-4e2c-9d1a-abcdef123456"
}
Example: Ingest Website Content 
curl --request POST \
  --url https://api.qontext.ai/v1/ingestion/website \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "workspaceId": "ws_123e4567-e89b-12d3-a456-426614174000",
  "knowledgeGraphId": "kg_681c0068-b589-4923-ac4a-73040fab7a45",
  "url": "https://gryffindor-inc.com/about",
  "type": "singlePage"
}'
Response: 
{
  "syncLogId": "sync_b5c7f1d2-3a4b-4e2c-9d1a-abcdef123456"
}