search

Semantic search identifies and retrieves semantically relevant memories by using a semantic index and in-memory embeddings to determine contextual relevance. This enables query results based on meaning rather than exact term matching across stored records.

The best way to find context for a briefing is querying your stored records to surface summaries and relevant context. Semantic retrieval operates across all mined memories, projects, and conversations to support planning and decision-making.

Yes, you can use semantic search to locate records from across your entire knowledge base. It operates across all stored memories, projects, and conversations, returning concise summaries or full-context results based on semantic relevance.

Yes, you need a semantic index and in-memory embeddings to perform contextual information retrieval. These components determine semantic relevance and enable the system to return concise summaries or full-context results for your queries.

You query stored memories by asking your system to locate the most relevant records for your topic. The semantic search mechanism then retrieves matching memories, projects, and conversations to inform your decision-making process.

A limitation of semantic search is its dependency on a pre-existing semantic index and in-memory embeddings. Without these, the system cannot determine contextual relevance or retrieve accurate summaries from stored conversations and projects.