Commit Graph
4 Commits
Author SHA1 Message Date
yhh 76ed2566ed feat(context): a fold archives the whole span and keeps nothing: no first-turn pin, recent tail, kept user turns, user-words list, work index or summary; the pointer is one short message and keep, recent_keep, user_turn_keep_tokens and first_turn_pin_tokens are retired 2026-11-06 10:02:32 -08:00
yhh 1b8b56aa65 feat(studio): compaction archives earlier turns to a file instead of summarizing; a skipped ask ends the turn as finished; chat overlays follow the design board
Compaction moves the folded span into <session>.outputs/fold-<hash>.md and leaves a pointer the model reads back with read_file. The summarizer path, its retries, bounds and tests are removed.

Answering an ask with no selection ends the turn normally instead of reporting a failure.

Find bar, overview and context cards, compact dialog, image preview, root-missing banner, toasts, side panel and prompt refine are aligned to the Chat-Tools board; the agent preset picker follows the Agent-Picker board.
2026-11-05 16:25:25 -08:00
yhh 985853ac03 fix(compaction): a failed summary's retry hold gives way before the window, and a code-less overflow refusal is recovered
A transient summary failure held the retry until the input grew by an
eighth of the trigger. Recorded near 105% of the trigger that release
point is the window itself, so the next request went out uncompacted.
The step is now capped at half of the room left below the hard ceiling,
so the hold always lifts before it.

A relay that refuses an oversized request with a 400 and no error code
was not recognised as an overflow, so recovery never ran. A 400 or 413
whose request our own estimate puts over the window now takes the same
fold-and-replay path, judged by status and size, never by wording.
2026-10-05 01:47:19 -07:00
esengine 7db5d0af31 fix(compaction): release a transient summary failure once the input has grown
A truncated, timed-out or failed summary request blocked automatic
compaction for the whole generation, so context kept growing past the
fold trigger with no further attempt. Block release now reads the typed
failure code: transient request failures are released after the input
grows by one retry step; every other class stays blocked.
2026-10-04 08:32:39 -07:00