Two new QuestionComposer cases: the collapse toggle hides the option list
and leaves the header strip, and collapsing between answering preserves
both the picked draft and the current question position through a full
submit; re-expanding does not steal focus back into the textarea. Refresh
the question-composer and steering web e2e goldens for the new header
button.
The ask-user takeover rendered the pending question set as a bottom card
capped at min(60vh, 520px) with no way to reduce it, which buried the
conversation above while the user decided. Add a minimize toggle next to
the dismiss action: collapsed, the card becomes a header strip (title plus
the two icon buttons) and the option body and footer unmount; expanding
restores the full card. Drafts and the current question index live in
QuestionFlow local state, so collapse/expand never loses them. The
free-form textarea autofocuses only on first presentation, so re-expanding
does not steal focus from the toggle.
Agent Note: .agents/notes/implemented/feature/2026-08-11-collapsible-ask-user-question-card.{md,zh.md,i18n.yaml}
A max-tokens response that included a tool call persisted assembler-transformed
content next to replay metadata projected from the untransformed native message,
so the next request died in history reconstruction with INVALID_REPLAY_STATE and
the session stayed permanently stuck.
Write side: the finish chunk's replayState becomes a typed ReplayEnvelope —
opaque response-level metadata plus optional per-block entries aligned with the
emitted block sequence. BlockAssembler computes one keep/drop decision for
blocks and entries together, so stored metadata always describes stored content
and retained blocks keep their signatures. pi-ai splits its state into a
version-2 response half and per-block signature entries.
Read side: durable content is authoritative. toPiAssistant degrades any
unusable state — foreign kind, other versions (including the flat v1 form
already on disk), malformed metadata, or content/block mismatches — to the
existing provider-neutral conversion with an onReplayDegrade diagnostic instead
of failing the request, which un-bricks sessions poisoned before this change.
Covered by assembler and replay unit tests, an agent-loop continuation
regression, keyless real-composition continuation tests (native pruned-envelope
replay and legacy flat-state degrade), and the authored keyless snapshot
scenario max-tokens-continue through the assembled ACP app.
tool-bash-persistent overwrote the backend's PS1, so terminal-bash prompt
readiness never matched and every send degraded to the 3.5s silence tier
(idleSilenceMs + handoffGraceMs) under production defaults.
The controlled PROMPT_COMMAND now re-asserts PS1 before every prompt, so an
in-shell override never survives to the next prompt. The tool initializes
with stty -echo alone and detects the no-end-marker fallback through the
seam's stdin_read wait reason instead of matching its own prompt text.
Tool calls drop from 7180/3560/3566 ms to 355/88/91 ms (spawn+init+echo,
echo, pwd; darwin, production defaults). The loader composition suite now
pins the fast path by pushing idleSilenceMs beyond the send bound, and a
real-PTY case proves PS1 self-healing.
Fixes#2585
Yichen Jiang