Effort, one generation at a time
How muxcode picks a seat's reasoning effort before every model call, with the numbers from its first live runs.
Read the noteField notes
Notes from the index.
Notes on running a coding agent: the harness on your machine, the context it keeps, the sandbox it works in, and the index it searches.
-
-
No compaction
When a task fills its context window, pi opens a fresh one with a short handoff instead of a summary; here is where that line sits, what crosses it and what it costs.
Read the note -
Plan first, then prove it
How a writing run in muxcode plans a change, builds it and proves it, and why a model's yes is never enough to pass the work.
Read the note -
How muxcode works
The route one change takes through muxcode, the questions every tool call is asked, and where the sandbox fits, with drawings you can step through.
Read the note -
How the sandbox is held in
A sandbox task runs your agent's commands on Cloudflare while the model, its keys and every permission stay on your Mac. These are the walls around it, and what each one stops.
Read the note