Illuminate

POST /illuminate goes beyond retrieval: it runs an agentic loop over the namespace's memories to answer higher-level questions, synthesize summaries, and surface patterns no single memory states.

Where search returns memories for your prompt, illuminate returns a conclusion — it searches, reasons, searches again if needed, and writes a synthesis.

Request shape

{
  "query": "what has the payments team decided this quarter?",
  "budget": "mid",
  "max_tokens": 1024
}

• budget — low for quick syntheses, high for deep multi-step reasoning.
• context — extra framing for the conclusion ("you are briefing a new hire").
• response_schema — a JSON Schema; illuminate returns structured output matching it instead of prose.
• tags, tags_match, tag_groups, fact_types — scope which memories the loop may draw on.
• include_facts — return the supporting facts alongside the conclusion.
• exclude_automations — leave standing automation outputs out of the evidence.

Namespace identity

Illuminate answers in light of the namespace's configured illuminate mission and disposition — set them via namespace config. A support namespace can answer cautiously and cite tickets; an engineering namespace can be terse and technical.

Automations

An automation is a standing illuminate query that re-runs as new memories land, so dashboards and briefings stay current without polling. Illuminate powers each refresh; results are queryable per automation.

Illuminate meters against the illuminate quota — see Rate limits.