Switch between Claude and Codex with a shared Obsidian memory
Johan Slåttavik can switch between AI agents such as Claude and Codex without explaining the project again. He keeps project history, documentation and decisions in a development memory in Obsidian that the agents can use.
Done with CodexModerate
What was done
Johan Slåttavik can switch between AI agents such as Claude and Codex without explaining the project again. He keeps project history, documentation and decisions in a development memory in Obsidian that the agents can use. The post does not show how the agents read and update the memory, or how much time it saves.
How to do it
- Paste the prompt into the tool, and replace the text in brackets with your own.
- Read the answer through. Do not send, delete or publish anything until you have seen it yourself.
- Check the original before you trust a figure, a name or a date.
Try it with Codex
Read [the project folder] and create a note in [the Obsidian vault] with three parts: what the project does, decisions we have made and what is left. Keep it short, with a date on each decision. Do not include passwords, keys or customer data. Update the note every time we end a session. You are done when another agent can read the note and say what the next step is.
Replace text in brackets with your own.
Done when. Another agent can read the note and say what the next step is.