Marrow MCP server
io.github.aryawidjaja/marrow·v0.8.0·3 starsShared memory for parallel AI coding agents. Local, free, with rooms and file claims.
3 stars
Reviews
Write oneNobody has reviewed Marrow yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Marrow tools
No tool declarations could be read from the package source. They show once the server is installed.
Public scan report
scanner v0.1.9 · 2026-09-23 · same rubric, same numbers if you re-run it
- –Code scanpackage could not be scannedn/a
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancelast push 43 days ago12/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How Marrow compares
Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 80/100
- higher than 60% of 48,255 servers
- GitHub stars
- 3
- more than 80% of 28,688 servers
Install Marrow in Claude Code, Cursor or VS Code
The registry entry has no remote endpoint or installable package.
What the publisher says
From the Marrow repository's README, as published. We do not edit it. Read it on GitHub
marrow
Memory that keeps working after Claude's runs out — shared across your projects, machines and tools.
Claude's memory stops at 25KB. This one doesn't.
Claude Code's built-in memory loads the first 200 lines of one file into every session. That is fine for a young project. A codebase two years old knows more than fits in 200 lines, and everything past the cut is simply dropped.
Marrow retrieves instead of loading. Ask a question and it returns the twenty memories that answer it, plus the ones linked to those, out of however many thousand you have.
At 1,000 project facts that is 2.1× less context per turn (p = 0.002) and $0.50 a task instead of $0.90. Numbers and method below.
It also does three things the built-in memory does not:
api recall what webapp knows.
- One brain across projects. Built-in memory is per repository. Marrow's hive lets an agent in
syncs through a relay you run, or Spinal Cloud if you would rather not.
- One brain across machines and teammates. Built-in memory is explicitly machine-local. Marrow
- One brain across tools. Claude Code, Cursor and Codex read and write the same memory over MCP.
And because several agents share it, it also keeps them from colliding: a file another live session is editing is claimed, and every agent's actions land in one append-only, hash-chained record you can read.
Free forever, AGPL-3.0, runs on your machine. Every memory is a markdown file you can open and delete.
Does it actually help? We measured it
The usual way to tell an agent how your project works is to write it all into a CLAUDE.md, which it then reads on every single turn. That is fine for ten things. A codebase a couple of years old knows a thousand.
We gave a coding agent the same task and the same repo three ways: nothing, everything in a CLAUDE.md, and the same facts in Marrow. 75 runs.
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Marrow: common questions
- Is Marrow MCP server safe?
- Mostly: it is graded B (80/100). Read the Marrow safety report
- Does Marrow need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Marrow maintained?
- The last commit was 84 days ago (2026-08-12). The latest release is v0.8.0.
- What can I use instead of Marrow?
- Servers from other publishers that do the same job: Deja Vu MCP server, Concord MCP server and Claudexor MCP server.
- How does Marrow compare with other MCP servers?
- safety score 80/100 (higher than 60% of 48,255 servers); github stars 3 (more than 80% of 28,688 servers). Full comparison.
Alternatives to Marrow
Same job from other publishers: the closest match first, then the best rated.
- Deja Vudeja-vu: local memory over the session histories of forty-three coding agents.not reviewedEstablishedA
- Concord MCPCross-harness communication and shared work-state for AI coding agents.not reviewedEstablishedA
- ClaudexorRoute coding work across Claude Code, Codex, Cursor, and OpenCode with shared context.not reviewedEstablishedA
