mcp-memory-service v11.9 — 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.

Try the Live Demo Codeberg PyPI Package

Interactive 3D knowledge graph — every memory a glowing node, every relationship a curved edge.

What's New in v11.9

MINOR release: transformers 5.x closes two high-severity advisories with no fix in the 4.x line, verified by a new CI job that finally installs the ml/nli extras — and that job caught a real quality-system bug on day one. ~2,966 tests.

🔒

transformers 5.x, Two Advisories Closed

GHSA-29pf-2h5f-8g72 and GHSA-fgcw-684q-jj6r are both fixed only in 5.x, so the [ml] and [nli] extras now require transformers 5.5 or later, below 6. The reason to upgrade if you install either extra. (#305)

🔧

Disabled Quality System Stopped Exporting Models

MCP_MEMORY_QUALITY_ENABLED=false still triggered a full torch.onnx.export of DeBERTa on the first call, because both entry points checked config.enabled after initializing. Found when it blew pytest's 120s CI timeout. (#316, closes #315)

📰

ML Extras Finally Run in CI

transformers and sentence-transformers had never been imported in CI before this job (#303) — it's what made the transformers 5.x move verifiable, and what caught the quality-system bug above. Cache-resolution and setuptools fixes ride along. (#307, #300, #310)

How It Works

Claude.ai connects to your memory server through Streamable HTTP with OAuth 2.1 authentication.

Browser
Claude.ai
→
Transport
Streamable HTTP
→
Auth
OAuth 2.1 + PKCE
→
Server
MCP Memory
→
Storage
Knowledge Graph
Data flow Auth boundary

Built for Production

Battle-tested with comprehensive testing and optimized for performance.

0
Tests
0
AI Apps Supported
0
Read Latency
0
Cache Speedup

In the Media

Talks and coverage from the community.

📰

Agents Overdrawn at the Memory Bank

Heavybit's Humans in the Loop deep dive talks to maintainer Heinrich Krupp about agent amnesia and why persistent memory is the missing infrastructure layer for agentic systems.
🎤

AI Tinkerers Zürich Talk

Maintainer Heinrich Krupp presents mcp-memory-service at the AI Tinkerers Zürich meetup — persistent memory architecture, semantic search, and multi-agent memory sharing. Watch the full video →

Get Started in Seconds

Install from PyPI and connect your AI agents to persistent memory.

Click to copy pip install mcp-memory-service
View on Codeberg Documentation Live Demo: Memory in Your Browser Blog: 5-Min claude.ai Setup Release Notes

Deploying for a team?

I help teams run self-hosted memory in production — OAuth, hybrid sync, multi-client rollout. Free 30-minute intro call.

Get in touch