Create source
Register a content source (url, text, or uploaded file) for ingestion. Returns 202: the source row commits immediately and extraction runs asynchronously in the worker; poll getSource or listSources for the extraction status. Paid extraction (e.g. TikTok fetches, embeddings) debits credits in the worker pipeline, not at this endpoint.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Path Parameters
Team (workspace) id (UUID). Discover ids via the list teams endpoint or the agent context.
Body
Kind of source to ingest: text, article, youtube, tiktok, audio, pdf, or skill. Determines which of url, text, or storagePath is required.
text, article, youtube, tiktok, audio, pdf, skill "article"
Public URL to extract from; required for article, youtube, and tiktok sources.
"https://example.com/blog/launch-post"
Raw text content to ingest directly; required for text sources.
Storage path returned by createSourceUploadUrl after uploading the file; required for audio and pdf sources.
Display title for the source; derived from the content when omitted.
"Q3 launch blog post"
Optional guidance for the extraction step, e.g. what to focus on or ignore.
"Focus on the pricing changes; skip the changelog."
Whether to keep the source permanently (true) or leave it as a retention candidate (default false).
Response
Success
text, article, youtube, tiktok, audio, pdf, skill pending, extracting, ready, failed, skipped_insufficient_credits