These goldens were recorded on a machine with personal skills under ~/.dsh and
~/.agents, so every one of them pinned a context-injection row a clean runner
cannot produce — which is why the browser lane failed on CI and passed
locally. master's scaffold fix confines skill-local's host-level roots to the
temp world; re-recording against it removes the row and nothing else.
seeded-history keeps its row: that scenario seeds the event itself.
Merging master swapped the plain "▸ 上下文注入" button for the shared
DisclosureRow, which every conversation golden carrying an injected-context
row renders. The only change in all twelve files is that row.
The Web composer now renders a request that declares the `plan-review` intent
as its own surface: the waiting-approval card language — amber "Plan review"
strip, the plan as the scrolling markdown body, the question as the card's
accessible name — over one decision row of Chat about it / Refuse / Approve.
Gone from that surface are the 1/1 pager, the numbered radio rows, the custom
answer row, and Skip/Submit, which made approving a plan read as sitting an
exam.
Approve and Refuse answer with the asker's own option labels and keep its
descriptions as tooltips; Chat about it cancels the request so the composer
returns and the user can simply say what they want. Copy is bilingual under the
existing `question` namespace.
The shape choice lives inside the single composer entry rather than a second
chain registration, so the two surfaces cannot race the same carrier, and
`planReviewOf` falls back to the generic flow for any request it cannot render
as a card — the client sits downstream of a wire boundary and every request
must stay answerable.
Review follow-ups on the approval takeover:
The scroll region is now a tab stop (tabIndex + named role="group"). The
question composer's scroll body needs none — its option rows are focusable and
pull the container along — but this one holds nothing but text, so a
keyboard-only user could reach the buttons and never the command's tail, and
approve what they could not finish reading.
The action row's padding reproduces the 14px gap it had inside the body: the
flex gap of 6 plus its 8px top margin, neither of which reaches it now that the
row sits outside the scroll region. The resting card is unchanged again.
The answered-transcript golden captured the OS's own refusal of the denied
first attempt — "bash: notes.txt: Operation not permitted" on macOS against
"bash: line 1: notes.txt: Read-only file system" on Linux — so it passed
locally and failed the Linux snapshot lane.
The scenario now keeps one golden (the waiting panel, platform-neutral) and
asserts the answered state on the world instead: the decided outcome, the file
the escalated command actually wrote, DONE, the panel gone, and the composer
re-enabled. The file assertion is stronger evidence than the transcript dump
it replaces — it proves the grant reached the executor.
The approval panel replaces the InputBar while a sandbox escalation waits,
and its justification and command are unbounded model text. With no height
cap, a long command grew the card until the refuse/allow row went under the
fold: at 900x700 the action row's bottom landed at y=749, so the user could
read the request and not answer it.
Justification and command now scroll in one region capped at the same height
as the composer's draft area, with the amber strip and the action row outside
it. The cap is one value with two consumers — declared as
--dsh-composer-text-max-height on ConversationRoot's .composerSeat, the
composer chain's only shared ancestor — so the seat cannot cap its two states
differently. The card rebinds the l2 scrollbar pair like every other scrolling
surface on an elevated background.
Covered by a new web e2e scenario that drives the real composition (read-only
session, denied write, the model's escalation retry, answer clicked through
the panel) and measures the live panel at two viewport heights against the
composer's own cap, read off the textarea rather than hardcoded.
The master-side refresh batch recorded these goldens against master's
composer (customTrigger row, header pager); the merge keeps this
branch's redesigned composer (always-visible custom input, footer
pager) with copy on the standard locale seat, so the waiting aria and
the answered transcript re-record against the merged UI. The answered
golden also picks up the ask_user_question toolview row this branch
introduces.
The English-locale golden refreshed on master (8c753fd8f1) was recorded
against master's pre-miller dialog; this branch's own golden predated the
miller two-pane landing and was never enforced before the web browser
snapshot CI gate landed. Re-recorded against the merged tree: two-pane
landing plus the show-hidden footer toggle.
The Models page drops the generic schema renderer and the visible
environment-variable field: each editor is a curated per-family card whose
primary input is one write-only API key stored under a derived
<ROUTE>_API_KEY reference (recorded as apiKeyEnv in the pi-ai profile), an
unkeyed whole-section provider opens as its setup card, and the collapsed
customized-settings fold carries baseURL/reasoningEffort (deepseek) or
reasoning (pi-ai). dsh-client-schema-form reduces to the schema/draft model
layer (no React).
apps/cli/cordis.yml gains settings-local, credentials-local, and the bare
dormant llm-pi-ai row (manifest deps added for the resolver contract);
llm-deepseek drops its !!js apiKey inline for per-request credential
resolution. Both adapters tag apiKeyEnv role('credential-ref') so the
form mounts the credential control. The web e2e scaffold isolates a
harness home per run — an in-process boot must never touch the
developer's real ~/.dsh — and the new models-settings scenario pins the
whole loop through the shipped app: dormant directory as add vocabulary,
schema-driven editor apply landing in settings.yaml, the route
registering live (topology frame), and a write-only key landing in the
temp .env with the configured badge converging. A hermetic test-owned
reference name keeps a developer's real provider keys from flipping the
badge. schema-form joins the platform module table (seed + externals)
so client bundles share one instance.
The seven behavior-pinning snapshots (workspace-flow, slash-flow,
todo-display, session-title, session-actions, terminal-card,
code-mode-fixture) asserted per-package behavior through the whole
built app: every assertion now lives in the owning package's suite at
equal strength (SlotTestRuntime benches assemble the same plugin apply
over the production slot machinery), while the assembled layer kept
re-failing on unrelated churn at 10s-timeout debugging cost.
What only this lane can prove is the artifact plane: lib/client.js
bundles loading through the ModuleLoader path, staged boot activation
across the inject layers, per-plugin CSS injection, and a journey
reaching chat content over the fixture transport. built-boot.snapshot.ts
pins exactly that, and nothing else — behavior assertions belong to the
package suites.
