REST API
WebSocket
Ingest a document
Ingests text content into a knowledge base namespace. For small text payloads the operation completes synchronously and the response includes chunks_created. For large content or file uploads (future) the response returns status “processing” and the caller polls /knowledge_base/get_status. Always returns a job_id for tracking. Requires ‘full’ scope.
curl --request POST \
--url https://api.svantic.com/knowledge_base/ingest \
--header 'Content-Type: application/json' \
--data '
{
"namespace": "string",
"source_id": "string",
"content": "string",
"tags": [
"string"
],
"source_type": "string",
"metadata": {}
}
'{
"ok": true,
"job_id": "string",
"status": "completed",
"async": true,
"chunks_created": 0,
"source_id": "string"
}Documentation Index
Fetch the complete documentation index at: https://docs.svantic.com/llms.txt
Use this file to discover all available pages before exploring further.
Body
Target knowledge base namespace.
Unique identifier for the source document (e.g. filename).
Text content to ingest (markdown, plain text, etc.).
Optional tags for filtering during retrieval.
Content type hint (e.g. "markdown", "text").
Arbitrary metadata attached to the source.
Response
Ingest completed or accepted for processing.
Job identifier for tracking (poll via /knowledge_base/get_status).
completed for sub-threshold sync ingests; processing when content length exceeds the async threshold (see gateway knowledge_base route).
completed, processing True when ingest continues in the background (large content).
Present when status is completed (sync path).
Echo of the ingested source identifier when completed.
curl --request POST \
--url https://api.svantic.com/knowledge_base/ingest \
--header 'Content-Type: application/json' \
--data '
{
"namespace": "string",
"source_id": "string",
"content": "string",
"tags": [
"string"
],
"source_type": "string",
"metadata": {}
}
'{
"ok": true,
"job_id": "string",
"status": "completed",
"async": true,
"chunks_created": 0,
"source_id": "string"
}