billy12151/memory-arbiter-mcp

Overview

Shared memory layer for AI coding tools. Uses one local SQLite database so every tool reads the same verified facts. Local-first with zero cloud dependencies.

Best for

Best for
Developers who want a local shared fact store across several AI coding tools.

Use cases

• Share a single source of facts across multiple AI coding assistants
• Keep tool memory consistent without cloud services
• Store and retrieve verified context locally during development

How to use

Install

pip install memory-arbiter-mcp

Tools exposed

• db_path
• backup_jsonl
• policy_path
• include_size

Tested with

Claude Desktop, Claude Code, Cursor, Continue

Notes

Shared memory layer for AI coding tools. Uses one local SQLite database so every tool reads the same verified facts. Local-first with zero cloud dependencies.

7 stars on GitHub. Last updated 2026-09-02. Licensed Apache-2.0.

Use cases

• Share a single source of facts across multiple AI coding assistants
• Keep tool memory consistent without cloud services
• Store and retrieve verified context locally during development

Pros

• Local-first design keeps data on your machine
• Simple SQLite storage is easy to inspect and back up
• One shared database avoids fragmented tool memory

Cons

• Very low adoption with only 7 GitHub stars
• SQLite may become a bottleneck for concurrent high-volume writes
• Fact verification process is not explained in the available description

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.