You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Nov 7, 2026. It is now read-only.
Repository navigation
This repository was archived by the owner on Nov 7, 2026. It is now read-only.
run: turn summary truncates from the right, hiding duration and the served model at narrow widths #13
Truncation is right-edge, so at narrow terminal widths the tail is silently cut. #12 makes the line longer by appending the model(s) that actually served the turn — Firerouter glm (glm-5p3-flash → glm-5p3) — which makes this reachable at common widths, especially with a multi-step route or a long raw served id.
Current priority order is defensible but accidental: duration is cut first, then the served parenthetical, then the configured model. The most valuable new information is cut last, which is the right order — but nothing enforces it, and nothing tests it.
Suggested fix
Reserve width for the duration, truncate the configured model label independently of the served parenthetical, and add rendering assertions at 60 and 80 columns.
It needs width measurement and a layout change to a shared scrollback component, beyond the display change that PR makes. Filed so it is tracked rather than lost in a PR comment.
Activity
fix(run): accumulate served models per turn and stop borrowing the se…
feat(tui): show the model that actually served a routed turn (#12)
fix(run): fit the turn summary to the width instead of cutting its ta…