Official Memvid plugins for Claude Code
npx claudepluginhub memvid/memvid-mindClaude Mind - Give Claude photographic memory in ONE portable file. Share, version, and transfer your Claude's brain.
memvidOne file. Instant recall. Zero config.
Installation • How it Works • Commands • FAQ
https://github.com/user-attachments/assets/b57cb3db-576b-4c1f-af92-95796ba3fb5b
You: "Hey Claude, remember when we fixed that auth bug?"
Claude: "I don't have memory of previous conversations."
You: "We literally spent 3 hours on this yesterday"
Claude: "I'd be happy to help you debug it from scratch!"
200K context window. Zero memory between sessions.
You're paying $200/month for a goldfish with a PhD.
You: "What did we decide about the auth system?"
Claude: "Last week we chose JWT over sessions because of your
microservices architecture. I also remember we had issues with
the refresh token rotation - here's what we fixed..."
One plugin. One file. Claude remembers everything.
30 seconds. No config.
# One-time setup (if you haven't used GitHub plugins before)
git config --global url."https://github.com/".insteadOf "[email protected]:"
Then in Claude Code:
/plugin marketplace add memvid/memvid-mind
/plugin install memvid-mind
Restart Claude Code. Done.
After install, Claude stores memories in one portable file:
your-project/
└── .claude/
└── mind.mv2 # Claude's brain. That's it.
| When | What |
|---|---|
| Session start | Injects relevant context from past sessions |
| While working | Decisions, bugs found, solutions, file changes |
| Session end | Session summary for future recall |
| Action | Result |
|---|---|
git commit | Version control Claude's memory |
scp file user@server: | Transfer to any machine |
| Send to teammate | Instant project onboarding |
No database. No background service. No API keys. No cloud.
/mind stats # memory statistics
/mind search "authentication" # find past context
/mind ask "why did we choose X?" # ask your memory
/mind recent # what happened lately
Claude hits context limits fast. memvid-mind compresses tool outputs ~20x:
Before: Read (8K) + Edit (4K) + Bash (12K) = 24K tokens gone
After: Read (400) + Edit (200) + Bash (600) = 1.2K tokens
Keeps errors, structure, key functions. Drops the noise.
Work longer without hitting limits.
The .mv2 format is designed for instant operations on files of any size. It pre-allocates:
This is the same architecture used by SQLite, LevelDB, and other production databases. The 4MB overhead enables sub-millisecond operations whether you have 10 memories or 10 million.
Think of it like an empty filing cabinet - it takes up space, but it's ready to organize thousands of documents instantly.
~1KB per memory. 1000 memories ≈ 1MB additional. The base 4MB is fixed overhead.
100% local. Nothing leaves your machine. No telemetry. No cloud sync. You own your data.
The .mv2 file is just a file - encrypt it, back it up, delete it, share it. Your choice.
Native Rust core via memvid. Sub-millisecond operations:
rm .claude/mind.mv2
Or use /mind clear (coming soon).
Yes! Each project gets its own .claude/mind.mv2 file. Memories are project-scoped by default.
Create .claude/mind.config.json:
{
"memoryPath": ".claude/mind.mv2",
"maxContextObservations": 20,
"endlessMode": true
}
Most users don't need to configure anything.
We love contributions! See CONTRIBUTING.md for guidelines.
Matt Pocock's skills for real engineering, as an installable Claude Code plugin.
Plugins by Alex Newman (thedotmack)
Ultra-compressed communication mode for Claude Code. Cuts 65% of output tokens (measured) while keeping full technical accuracy.
'%20stop-opacity%3D'0.16'%2F%3E%3Cstop%20offset%3D'1'%20stop-color%3D'rgb(200%2C90%2C60)'%20stop-opacity%3D'0.03'%2F%3E%3C%2FlinearGradient%3E%3C%2Fdefs%3E%3Crect%20width%3D'320'%20height%3D'200'%20fill%3D'url(%23g)'%2F%3E%3Ccircle%20cx%3D'250'%20cy%3D'56'%20r%3D'92'%20fill%3D'rgb(200%2C90%2C60)'%20fill-opacity%3D'0.06'%2F%3E%3Ccircle%20cx%3D'64'%20cy%3D'172'%20r%3D'58'%20fill%3D'rgb(200%2C90%2C60)'%20fill-opacity%3D'0.05'%2F%3E%3C%2Fsvg%3E)