memshare
MIT MCP server and CLI that keeps an assistant memory store locally and exports reviewed bundles a teammate can import.
0 members list memshare, unchanged over the last 12 weeks, data as of 16 October 2026
Members who list memshare in their Stack, by week.
Be the first to add memshare
Nobody on Stackness runs memshare yet. Add it to your stack and you're the first member running it.
About memshare
Memshare is an open source tool that stores AI assistant memory locally as JSON files and enables peer-to-peer sharing between users with explicit consent. It functions as a Model Context Protocol server and command-line interface, allowing developers to manage memories across different AI models and tools without vendor lock-in. The tool treats memory as data files rather than platform-specific features, making stored context portable and inspectable.
Memshare runs on a user's own machine with no central server, so each person controls their memory data independently. Users can capture memories automatically as they work with AI assistants, review what gets saved, and selectively export bundles for teammates to import into their own local stores. The project is developed by kampana and licensed under the MIT license, written in TypeScript.
Checked 16 October 2026.
Nobody on Stackness runs memshare yet.
Sign up and add itPosts about memshare
- Agent context compression tools: Headroom, RTK, Engram, Attic and how each one measured its saving
Seven tools sit between a coding agent and the model as of October 2026. Every headline saving is self-measured, and RTK is the safest place to start.