mcp-memory-service v10.50 — 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 v10.50 — 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.
Plugin hook scaffolding — extend memory without forking core. ~1,838 tests.
Four lifecycle hooks — on_store, on_delete, on_retrieve, on_consolidate — let third-party packages inject logic at key points using entry_points discovery. No core fork required. (PR #856, @filhocf, refs #732)
Plugins register via standard Python packaging (entry_points group mcp_memory_service.plugins). Install a plugin package and hooks activate automatically — no config changes needed. Enables the ecosystem extensibility milestone.
Latest authlib, cryptography, and torch releases. CI actions updated: attest-build-provenance, codeql-action, and hadolint-action all bumped to current majors. All 905 integration tests pass. (PRs #858–#861)
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.