Find the decision behind the answer.
A question about past work often needs more than a matching sentence. The query skill guides your assistant through search, relevant pages and recorded relationships so its answer can point back to the evidence.
Illustrative request
Why did we choose this architecture, and did a later review change the decision?
The assistant searches decision notes and related timelines, then cites the evidence it found. Conflicting or missing records stay explicit; the workflow cannot recover a decision that was never saved.
Illustrative scenario, not a live product result.Search in more than one way
Names and dates benefit from exact terms. Broader questions benefit from meaning-based retrieval. Relationship questions can use typed links and graph traversal. The skill chooses between these approaches and reads fuller pages when excerpts do not provide enough context.
Explain disagreement instead of hiding it
A new decision and an older meeting note may disagree for a good reason. The workflow asks your assistant to compare source authority and dates, retain both references where needed, and flag stale evidence. A few search hits are not proof that every relevant source has been found.
- Source pages behind factual claims
- Dates and context for changing decisions
- Visible gaps when the evidence is thin
Use the knowledge your credential can read
The skill runs in your connected assistant and is read-only. Memrain supplies retrieved material; the assistant composes the response. Available sources and tools depend on your credential. Optional query expansion uses model calls when enabled, so broader retrieval can have an additional usage cost.
Keep the context. Move the work forward.
Connect your knowledge to the AI you already use. Find your starting point and make the next conversation count.