Performs a semantic similarity search across a board’s vectorized document chunks. Returns the most relevant chunks ranked by similarity score.
Pass your API key (prefixed cgk_) or a Supabase JWT in the
Authorization: Bearer <token> header.
Board UUID.
"a1b2c3d4-e5f6-7890-abcd-ef1234567890"
Natural language search query.
1 - 2000"What were the Q1 revenue numbers?"
Maximum number of results to return.
1 <= x <= 50Minimum similarity score threshold (0-1).
0 <= x <= 1Restrict search to chunks associated with these nodes.
["e5f6a7b8-c9d0-1234-efab-345678901234"]