Claude Brain
A Claude Code plugin that keeps memory across sessions in one local file, with no database, cloud or API keys
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- Stores and reads project context and injects it at session start
- Data stays local in the project file
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/claude-brainDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Run one by one in the Claude Code chat
/plugin marketplace add memvid/claude-brain
/plugin install mind@memvidInstall the plugin in Claude Code: /plugin add marketplace memvid/claude-brain, then enable the mind plugin and restart the session. Use /mind search to query the memory.
Other ways from the author
/plugin add marketplace memvid/claude-brainThen enable the mind plugin in /plugins and restart the session.
This is third-party code. Review the repository files before installing.
What it does
Claude Brain gives Claude Code memory across sessions. After install the project memory lives in a single .claude/mind.mv2 file that captures session context, decisions, bugs and their fixes. The context is auto-injected at the start of a new session and is searchable. Because it is one file, it is easy to commit to git, copy over scp or send to a teammate for quick onboarding. Commands like /mind stats, /mind search and /mind ask work inside Claude Code, and a separate CLI gives direct access to the file. All data stays on your machine.
Who it is for. For developers on Claude Code who want project memory across sessions without external services.
Good fit when
- You want Claude to remember decisions and bugs from past sessions
- You want to version the project memory in git alongside the code
- You want to onboard a teammate quickly with a single file
Not a fit when
- You need shared cloud memory synced across devices
- You are not working in Claude Code or a similar agent
Example request
What did we decide about auth last time, check the memoryLimitations
Memory lives in a single local file per project; there is no shared cloud store or cross-machine sync. If you have not used GitHub plugins before, a one-time git setup is needed. After install the mind plugin is enabled and the agent restarted. The optional CLI installs separately via npm.
How to disable. Disable the mind plugin in /plugins and, if needed, remove the .claude/mind.mv2 memory file in the project.
Security check
- Stores and reads project context and injects it at session start
- Data stays local in the project file
README in short
The README presents Claude Brain as a fix for the lack of memory between Claude Code sessions. Project memory is stored in a single .claude/mind.mv2 file with no database, cloud or API keys, auto-injected at session start and searchable. Installation adds the marketplace and enables the mind plugin with a restart. Inside the agent the /mind stats, search, ask and recent commands work, and an optional memvid CLI gives direct file access. The core is in Rust and the project builds on the memvid memory engine. MIT licensed.
FAQ
Does data go to the cloud?
No. Memory is fully local, stored in the .claude/mind.mv2 file, and never leaves your machine.
How large does the file get?
Per the README an empty file is about 70 KB and grows roughly 1 KB per memory; a year of use usually stays under 5 MB.
Related
A self-hosted knowledge base with block-level references and a built-in MCP server for connecting AI agents to your notes
A CLI for every Google Workspace API with JSON output and agent skills: Drive, Gmail, Calendar, Sheets and more
Local search over Markdown notes, docs and meeting transcripts: keywords, semantic search and reranking, with an MCP server
A task manager for AI-driven development: breaks a PRD into dependent tasks and guides the agent through them via MCP or CLI