One thing that always bothered me about working with Claude is that every session starts from zero.
You explain your project, your conventions, your decisions. You get great help. Then the session ends. Next time you open a new chat, all of that context is gone and you have to explain everything again.
I use Logseq for personal notes and project tracking. At some point I thought: what if Claude could just read and write to my Logseq graph? That is how LogseqBrain was born.
What is LogseqBrain
LogseqBrain is a free and open source Claude plugin that turns a Logseq graph into persistent memory. Claude can read from and write to the graph, storing project plans, decisions, implementation details, and session notes.
Because Logseq syncs across devices, the context travels with you. Start something on your desktop, continue on your laptop, and Claude already knows where you left off.
Repo: github.com/jame581/LogseqBrain
How to Install
Claude Code:
/plugin marketplace add jame581/skillsmith
/plugin install logseq-brain@skillsmithGemini CLI:
gemini extensions install https://github.com/jame581/LogseqBrainAfter installing, create a Logseq graph called ClaudeBrain, point the plugin to it, and say "init brain" to set everything up.
How It Works
There are four simple commands:
init brain - sets up the graph structure and adds your first project.
load MyProject - loads your project context into the current session. Claude immediately knows your decisions, plans, and current focus.
save to brain - saves everything from the current session back to the graph. Decisions, progress, plans. It also detects Jira task context automatically if you are working on a task.
brain status - shows a quick dashboard of all your projects with current status and last activity.
That is the whole workflow. Load at the start, save at the end.
It is Free and Open Source
LogseqBrain is MIT licensed. If you use Claude and Logseq together, give it a try.
If it saves you time, you can support the work here.
GitHub: github.com/jame581/LogseqBrain