- Blend BM25/ts_rank relevance with importance instead of sorting by one
dimension only. Default mode: 40% relevance + 60% importance. Relevance
mode: 70% relevance + 30% importance.
- Try AND-match first for precise results, fall back to OR-match when too
few results are found. Prevents single-word matches from flooding results.
- Applied to both SQLite (local) and PostgreSQL (API) search paths.
- Add MAX_MEMORY_CHARS=800 Pydantic validation on MemoryStore.content
- Update auto-learn judge prompts: "ONE topic per event", 100-500 chars,
include the WHY not just the WHAT
- Split 9 mega-memories (800-2400ch) into 70 focused memories (100-500ch)
via migration script
Before: median 331ch, 11 memories >800ch, recall wastes 84% of returned tokens
After: median 213ch, 2 memories >800ch (dense single-topic refs), recall returns
only the relevant knowledge
Trade-off research: PostgreSQL ts_rank gives the same score regardless of
document size, so a 2400-char memory with 12 topics gets recalled for any
of its 12 topics but wastes context with the other 11. Focused memories
(100-500ch) give higher signal-to-noise per recall.
Old sync clients without 404 handling get stuck in infinite retry
loops when trying to delete an already-deleted memory. Making the
endpoint idempotent (returning success regardless) fixes this for
all existing clients without requiring client upgrades.
Three fixes for high 4xx alert rate:
1. Catch urllib.error.HTTPError (not just RuntimeError) for 404 on
DELETE in pending_ops — was causing infinite retry loop for
already-deleted memories
2. try_sync_delete: treat 404 as success instead of enqueuing
a retry that will also 404 forever
3. URL-encode the `since` query param to prevent `+` in timezone
offset being decoded to a space (the asyncpg-string-timestamp
pattern applied to the sync client)
The '+' in '+00:00' timezone offsets gets URL-decoded to a space,
causing datetime.fromisoformat() to fail with ValueError. Replace
spaces back to '+' before parsing.
Consolidates plugin components (hooks, commands, skills, MCP config)
into this repo so it can be installed with:
claude plugins install github:ViktorBarzin/claude-memory-mcp
- .claude-plugin/plugin.json: manifest with all hook events
- mcp/memory-mcp.json: MCP server config using existing src/
- hooks/: compaction survival, auto-recall, auto-learn, auto-approve
- commands/: /remember and /recall slash commands
- skills/: memory-management skill
- Bump MCP server to v2.0.0 with metaclaw migration fallback
- Update README with quick install and plugin hooks docs
- Add SyncEngine for background sync between local SQLite cache and
remote API with pending_ops queue for offline resilience
- Refactor MCP server to support three modes: SQLite-only, hybrid
(local cache + sync, new default), and HTTP-only (legacy)
- Add GET /api/memories/sync endpoint for incremental sync
- Change DELETE to soft delete (set deleted_at) for sync support
- Add deleted_at IS NULL filters to all read queries
- Scale API deployment to 2 replicas with pod anti-affinity, PDB,
and startup probe for high availability
- Add migration 003 for deleted_at column and updated_at index
- Add comprehensive tests for sync engine and API sync endpoint
- API recall and list endpoints now return {"memories": [...]} matching
the format expected by the MCP server
- Rewrote README with comprehensive setup instructions for new agents,
accurate API reference, migration docs, and multi-user setup guide
Replace inline migration logic with proper Alembic migrations:
- 001: Initial schema (creates memories table with FTS)
- 002: Add multi-user and secrets columns (user_id, is_sensitive,
vault_path, encrypted_content)
Migrations run automatically on app startup. Existing databases
are handled gracefully with IF NOT EXISTS / column existence checks.
