mcp-memory-service v10.63 — 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.

GitHub PyPI Package

What's New in v10.63

Milvus Issue #888 fully complete + Kiro CLI harvest fix with 36x parse yield improvement. ~1,985 tests.

🔍

Milvus Issue #888 Complete

The final 4 low-priority Milvus overrides are now native: search_by_tag_chronological (tag filter + server-side sort + pagination), count_memories_by_tag (Milvus count(*)), is_deleted (metadata check), and purge_deleted (hard-delete tombstones). All 9 optional BaseStorage methods now have native implementations, eliminating all base-class fallbacks.

🛠

Kiro CLI Harvest Fix

Kiro CLI uses AssistantMessage as the message kind (not Response). The harvest parser now maps this correctly and removes a redundant 2000-char filter. Parse yield for Kiro sessions improves 36x: 1 candidate per 71 messages to 36 candidates per 373 messages.

🔄

Community Contributions

Both improvements in this release come from community contributors: @henry201605 (Milvus overrides, PR #978) and @filhocf (Kiro CLI harvest fix, PR #979). Together they complete a major milestone for Milvus users and significantly improve Kiro IDE integration.

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

Get Started in Seconds

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

Click to copy pip install mcp-memory-service