Skip to main content
Your AI agent forgets everything between sessions. PMB remembers - locally.
PMB gives Claude Code, Cursor, Codex and other MCP agents a real memory: decisions, lessons, and project facts. They live in one SQLite file on your disk and come back through MCP - no cloud, no API keys, no LLM call on the read path. And it’s the only agent memory that tells you when it’s actually helping, instead of claiming “+X%”.
Without memory the agent asks clarifying questions; with PMB it recalls the answer in 6ms.
pip install pmb-ai
pmb connect claude-code

​
Start here

Quickstart

Install, wire your agent, and let memory work automatically.

How it works

Hybrid recall, async writes, hooks, and the storage model.

Does it help?

How PMB measures its own value - honestly.

CLI reference

Every command, by category: connect, recall, manage, hooks, config.

​
Why PMB

Local-first

One file on disk. No keys, no telemetry, no cloud to trust.

MCP-native

Claude Code, Cursor, Codex, Windsurf, Zed - one pmb connect.

Multilingual

50+ languages out of the box; a Russian query finds an English fact.

Honestly measured

Tells you when a memory isn’t helping. No fake numbers.

​
What it feels like

you  ▸ fix the pricing bug from last Tuesday
ai   ▸ which file is that in? what was the bug?
The agent didn’t ask. It read its memory and showed up to the task informed.