What’s New in MemHouse v0.4.0
MemHouse v0.4.0 brings dedicated architectural support for personal agents and ephemeral multi-agent swarms.
1. External State Memory for Ephemeral Swarms
When building multi-agent systems with LangGraph, CrewAI, or AutoGen, worker agents spawn for short subtasks and terminate. MemHouse serves as the external state engine that survives subagent process death:
- Subagents post observations to MemHouse before shutting down.
- Subsequent agents instantly retrieve distilled, governed facts without context window bloat.
2. Native Model Context Protocol (MCP) Server
Plugs directly into Claude Desktop, Cursor, OpenClaw, and LLM agent toolchains out of the box via @memhouse/mcp-server.
3. Personal Agent Connectors (OpenClaw & Hermes)
- OpenClaw CLI: Save terminal workflows, project configurations, and habits across system restarts.
- Hermes Agent: Python memory provider for local companion state with hard privacy scoping.
4. Automated Release Artifacts
This post is served directly from an Astro-compatible Markdown file (src/content/blog/release-v1-4-0.md) in the official repository.