mcp-memory-service v10.56 — Now Available
Memory for AI Agents
Persistent memory for AI agents — REST API, MCP, OAuth, CLI, dashboard. Semantic search, knowledge graph, automatic consolidation. One self-hosted service, every transport.
mcp-memory-service v10.56 — Now Available
Persistent memory for AI agents — REST API, MCP, OAuth, CLI, dashboard. Semantic search, knowledge graph, automatic consolidation. One self-hosted service, every transport.
Configurable maintain scan limit and smarter Insight Cards. ~1,902 tests.
New MCP_MAINTAIN_SCAN_LIMIT env var (default: 2000, set to 0 for unlimited) replaces the hardcoded 500-memory cap in the maintain cycle. Large deployments can now tune or remove the ceiling for entity extraction (Step 5) and insight card generation (Step 6).
InsightGenerator._detect_gaps now skips operational metadata tags (conflict:unresolved, automated, __test__, temporary, processed, auto-generated, insight-card) that are not knowledge domains, eliminating false-positive "Decision gap" insight cards.
Built on v10.55: EntityExtractor indexes @mentions, #tags, URLs, and file paths from memory content. InsightGenerator surfaces patterns, trends, and gaps via memory_quality maintain. Opt-in via MCP_INSIGHT_CARDS_ENABLED=true.
Claude.ai connects to your memory server through Streamable HTTP with OAuth 2.1 authentication.
Battle-tested with comprehensive testing and optimized for performance.
Install from PyPI and connect your AI agents to persistent memory.