mcp-memory-service v11.4 — 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 v11.4 — 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.
Interactive 3D knowledge graph — every memory a glowing node, every relationship a curved edge.
MINOR release: memory merge action, pluggable domain-specific NER extractors, and the new mcpmemory.services landing page with an in-browser demo. ~2,550 tests.
memory_consolidate gains a merge action: combine N source memories into one, with tag union, memory_type inheritance, store-first/delete-after safety, and honest partial-failure reporting. (#100, PR #105, @filhocf)
Extend entity extraction with your own domain extractors — medical, legal, technical. Implement the DomainExtractor protocol, register via MCP_ENTITY_EXTRACTOR_MODULES, done. Lazy-loaded, failure-isolated, deduplicated against built-in patterns. (#54, PR #107, @filhocf)
The new mcpmemory.services landing page ships a fully client-side semantic memory demo — sqlite-vec compiled to WASM plus transformers.js embeddings, running entirely in your browser. Plus a galaxy hero video rendered from 3,134 connected memories. (PR #106)
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.