lbb_ask
Ask natural-language questions about a knowledge graph and receive grounded answers with citations. Set 'execute: false' to retrieve only the mapped vocabulary without retrieval.
Instructions
Ask a natural-language question about the graph and get a grounded answer with citations. The database snaps the question to its real vocabulary (never invented), retrieves against the pinned snapshot, and answers. mode is resident_planner when a small resident model synthesized the prose, or grounding_only when it returns the grounded evidence for your own model to finish. Prefer this over lbb_search when you want a direct, cited answer rather than a ranked list; set execute: false to get only the grounding — the real vocabulary the question maps to — without retrieval. The response explain block reports how much the database narrowed (vocabulary candidates the question snapped to, plus retrieved entity/assertion counts) and the per-stage latency (ground / retrieve / synth / total ms), so you can see the pipeline that produced the answer.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| as_of | No | Valid-time cursor (RFC 3339): retrieval reflects facts true at this instant | |
| graph | No | Graph to target; defaults to the connection's graph | |
| top_k | No | Max citations to return (default 8) | |
| branch | No | Branch to target; defaults to the connection's branch | |
| detail | No | Response detail level. Defaults to compact. | |
| execute | No | Run retrieval and answer (default true); false returns only the grounding | |
| question | Yes | The natural-language question to answer from the graph | |
| as_of_commit_seq | No | Snapshot pin: retrieval and citations reproduce the graph as of this commit sequence |