Memory is the storage layer that keeps the context you have approved.
Memory acts as the database for your provider. It stores:
- Your approved context (the facts you accept or edit).
- The evidence chains showing which apps suggested what data and when.
- The age and decay status of each fact.
- Your review history (what you accepted, modified, or rejected).
When apps request your context through CAP, Memory helps resolve conflicting information:
- Evidence Weighting: Calculates an overall confidence score (0.0 to 1.0) using source trust levels and record age.
- Freshness: Tracks when data is current, aging, stale, or expired.
- Contradiction Resolution: Compares contradicting app suggestions, ranks them based on evidence, and returns the strongest active value.
Memory never automatically accepts app suggestions. Everything stays pending until you decide what to do in Notebook.
You can set records to:
pending: Suggestion from an app waiting for your review.approved: Accepted by you.user_verified: Confirmed or edited by you.rejected: Declined (retained only for logs).forgotten: Deleted at your request.
You can export your approved context at any time, allowing you to move to a different provider without losing your history.
Apache 2.0. Memory is open and free.