Interesting. I will have to give it a try. I have found that markdown costs me 40% more and takes much longer to search through and doesn't cross project boundaries.
Try out empirical if you can and let me know what you think.
No it's a graph memory storage and retrieval system usable by all AI via CLI or MCP tool.
Markdown as memory takes up too much context. Empirical loads dynamically via querying the context it needs on demand.
But neither travels cleanly across everything I use, and packing too much into MD files eats context and tokens.
With Empirical, I keep my AGENTS.md lean and let Codex pull context dynamically when it actually needs it.
I can open ChatGPT on my phone, connected to Empirical, and it pulls the same memory context and writing tone I use in Codex or any other connected AI tool.
That means:
* less repeated setup
* cleaner, cheaper prompts
* more consistent output across sessions