Search through stored Claude conversations for relevant content

Add a new conversation to the memory system.

session_id, user_id, tags, and conversation_type are the universal metadata fields introduced in PR #114; when provided, they are persisted alongside the conversation and indexed for metadata search (search_by_tag / search_by_session_id / search_by_conversation_type). All four are validated/sanitized before storage since this data may originate from external imports.

Update fields on an existing conversation in place.

Pass conversation_id plus any subset of fields to change. The first line of the stored content is rewritten with a self-documenting audit line: [update <iso-timestamp> — <change_note>]. If change_note is omitted, it's auto-derived from which fields changed.

Tag ops: set_tags replaces the full list; add_tags and remove_tags mutate it. set_tags is mutually exclusive with the other two; pass set_tags=[] to clear all tags.

Metadata fields (tags, conversation_type, session_id, user_id) are validated/sanitized before storage since this data may originate from external imports.

record_audit=False preserves authoritative imported content verbatim. Normal interactive updates should retain the default audit record.

Retrieve a stored conversation by an ID returned from a search tool.

The complete record is read from the authoritative JSON store. Content is capped to protect the model context; increase max_chars when necessary.

Generate a summary of conversations from the past week

Search conversations by a specific topic

Search conversations tagged with a specific tag (D2 metadata field).

Tags are universal metadata populated by the importers (e.g. starred, archived, workspace:my-project, variant:web). Requires SQLite FTS.

Find all conversations sharing a session_id (D2 metadata field).

Useful for reconstructing a multi-turn session that spans several stored conversation records (e.g. a Cursor working session, a Claude thread continued across days). Results are sorted chronologically. Requires SQLite FTS.

Search conversations by conversation_type (D2 metadata field).

Typical values: chat, code, analysis. Requires SQLite FTS.

Get search engine statistics and performance information