A LangGraph Note the Next Run Can Read

mountable.io is a persistent filesystem for this LangGraph example. The next node reads a note from the filesystem, and a later run with a new in-memory checkpoint reads it too.

This example stores a LangGraph note in an in-memory checkpoint. Then the process stops.

The next run starts with a new in-memory checkpoint. The note is gone.

mountable.io is a persistent filesystem for AI agents. The graph stores the note there. A later run can still read that file.

The next node reads the file

One graph has a write node and a read node. The writer stores the note on mountable.io and returns empty text.

The next node receives no note body in its input. It reads the file from the filesystem.

A later run reads it too

A later process starts with a new in-memory checkpoint. It opens the same filesystem read-only and reads the same path. The note is still there.

Your backend chooses the filesystem and the path before the graph starts. Use that filesystem id in the commands below.

Run the two commands

The LangGraph guide covers the account, the key and the filesystem. From the example:

uv run python graph.py handoff \
  --filesystem "$FS_ID" \
  --path notes/brief.md \
  --text "Parser review complete. Check empty input next."
 
uv run python graph.py resume \
  --filesystem "$FS_ID" \
  --path notes/brief.md

The first command writes the note and reads it back inside one graph. The second command is a new process. It prints the same note.

We maintain this community example for mountable.io. It uses your existing account. The commands talk to mountable.io only.