luxe 0.1.4: remembered models, wheel-driven lists, and a steadier phone app

- luxe starts on the model you last chose in this project, then the last one anywhere. Signing in
  opens the model picker instead of choosing for you.
- Every list in a modal takes the mouse wheel and a draggable scrollbar.
- The phone app shows reasoning in full, stays put while you read back through a long session, and
  shows a question's answer on its card.
- On Windows, luxe update replaces the running luxe, and a signed-in install no longer sees setup
  again.
- The file tree no longer shows "engine busy" in a large project, loop output is easier to follow,
  and the name is written lowercase.

The CHANGELOG has the full list.
This commit is contained in:
Evan Whittaker
2026-10-20 13:51:34 -05:00
parent 6184aa21fb
commit a0666d6c81
591 changed files with 4926 additions and 2942 deletions

View File

@@ -2,7 +2,7 @@
title: "Autonomous runs"
summary: "Let the agent work through a phased plan on its own, with tests and a reviewer at each phase."
order: 12
group: "Using Luxe"
group: "Using luxe"
---
`luxe loop` runs the agent on its own, phase by phase through your plan. A phase only counts as done
@@ -55,7 +55,7 @@ A run uses your session's model, or `--model`. Only one run at a time works in a
## Check each phase
When the agent says a phase is done, Luxe checks it before moving on:
When the agent says a phase is done, luxe checks it before moving on:
1. **Your commands.** Each must pass.
2. **A reviewer**, if `gate.review` is on and the commands passed. It runs in a fresh session, so
@@ -98,13 +98,22 @@ phase with `luxe loop phase --set <next>`.
A run prints one line per step, timed from its start:
```text
19s write logstats.py
49s test python3 -m unittest -q
1m09s phase 0 claimed complete — running the gate
1m31s the acceptance commands passed — asking the reviewer
2m34s phase 0 accepted — now on phase 1
0s ┃ phase 0
0s ┃ ┌ start
19s ┃ │ ✓ write logstats.py
49s ┃ │ ✗ test python3 -m unittest -q · failed
1m09s ┃ │ phase 0 claimed complete — running the gate
1m31s ┃ │ the acceptance commands passed — asking the reviewer
2m34s ┃ │ phase 0 accepted — now on phase 1
2m40s ┃ └ turn ended · 6 tools · 2m40s
2m41s ┃ phase 1
2m41s ┃ ┌ iteration 1
```
The heavy rail on the left marks the phase, in a color of its own, and the bracket inside it marks
each iteration. In a terminal, a tool's line shows `○` while it runs, then turns green with `✓` or
red with `✗` when it ends. Piped to a file there's no color, and a failure gets a line of its own.
`luxe loop status` shows what it has cost, by model:
```text