A catalog of 28 portable Agent Skills that give Claude Code, Copilot, Codex and Cursor structured reasoning procedures for decisions, diagnosis and risk.
memvid/claude-brain
A Claude Code plugin that gives Claude persistent memory stored in a single portable .mv2 file, with no database required.
- Stars
- 579
- Forks
- 59
- Open issues
- 10
- Last commit
- 9 months ago
What is it?
Claude Brain is a plugin for Claude Code that keeps session context, decisions, bugs and their solutions between conversations. Memory is stored in a single file at .claude/mind.mv2 inside the project, so there is no database, cloud service or API key to set up.
Because the memory is one file, it can be committed to git for version control, copied with scp, or sent to a teammate to share context. Stored memories are injected automatically when a session starts and can be searched at any time with the /mind commands, such as stats, search, ask and recent. A companion memvid-cli npm package offers direct command-line access to the file for searching, asking questions and viewing a timeline.
The core is built on the memvid engine with a native Rust implementation. The project describes the file as growing about 1KB per memory and searches as sub-millisecond.
Key features
- 01Stores all Claude Code memory in one .mv2 file inside the project.
- 02Automatically injects stored context at session start.
- 03Provides /mind commands for stats, search, ask and recent activity.
- 04Includes an optional memvid-cli for command-line queries and timelines.
- 05Runs entirely locally with no database or API keys.
Use cases
- Keeping architectural decisions and past bug fixes available across Claude Code sessions.
- Versioning Claude's project memory alongside code in git.
- Onboarding a teammate by sharing the memory file.
- Searching past debugging sessions to avoid repeating work.
Quick start
- 1
Set up git to use HTTPS for GitHub plugins
git config --global url."https://github.com/".insteadOf "git@github.com:" - 2
Add the marketplace in Claude Code
/plugin add marketplace memvid/claude-brain - 3
Open /plugins, go to Installed, enable the mind plugin, then restart Claude Code
Things to know
- Requires Claude Code; it is a plugin for that tool only.
- The README does not state a stable release or versioning, so expect possible changes.
- Memory search quality and behavior are not benchmarked in the README.
Topics
FAQ
- What is claude-brain?
- A Claude Code plugin that gives Claude persistent memory stored in a single portable .mv2 file, with no database required. Claude Brain is a plugin for Claude Code that keeps session context, decisions, bugs and their solutions between conversations. Memory is stored in a single file at .claude/mind.mv2 inside the project, so there is no database, cloud service or API key to set up.
- Is claude-brain free to use? What license does it have?
- claude-brain is open source under the MIT license. Check the license terms before using it in commercial products.
- Is claude-brain still maintained?
- No commits for 6+ months. The last commit to claude-brain was on Jan 19, 2026.
- What can I use claude-brain for?
- Keeping architectural decisions and past bug fixes available across Claude Code sessions. Versioning Claude's project memory alongside code in git. Onboarding a teammate by sharing the memory file. Searching past debugging sessions to avoid repeating work.
- How do I install claude-brain?
- Set up git to use HTTPS for GitHub plugins: git config --global url."https://github.com/".insteadOf "git@github.com:" Add the marketplace in Claude Code: /plugin add marketplace memvid/claude-brain Open /plugins, go to Installed, enable the mind plugin, then restart Claude Code
- Where can I get claude-brain?
- The source code and installation instructions are on GitHub: https://github.com/memvid/claude-brain
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A plugin-based harness that gives Claude Code, Codex and other coding agents shared agents, skills, hooks, memory and security scanning.
CLI proxy that filters and compresses shell command output before it reaches an AI coding agent's context, cutting token use.
A Claude Code plugin that adds a live status line showing context usage, rate limits, active tools, running agents, and todo progress.
A port of Poteto's pstack Cursor skill stack, giving Claude Code, Codex, Copilot, Pi and other agent harnesses structured workflows for coding tasks.
A macOS menu bar app that shows live Claude Code status: an animated icon while working, a yellow dot when awaiting permission, and elapsed turn time.