**A hand-declared route must not offer a reasoning effort.** The earlier
commit read the create card's missing control as drift and added one. It
is the other way round: such a model has no reasoning capability — pi-ai's
installed catalog is what supplies one, and it ships nothing under the
route — so `resolveModel` throws UNSUPPORTED_REASONING_EFFORT for every
model on it and the whole provider drops out of the picker. Verified
against the adapter, not inferred. The create card no longer offers it and
the editor withholds it on the directory's `declared` bit, which is the
real bug: that control has always been wrong for these routes.
**A blocked composer locked the way out of the block.** Reusing the
no-workspace inert posture disabled the model seat along with everything
else, so the bar asked for a model while preventing the one control that
picks one. A block now rides its own `blocked` owner prop: the textarea,
send, commands, plan seat, and access chip all lock, and the model seat
alone stays live.
**A Provider ID could derive an illegal credential reference.** The card
accepted a digit-leading id, whose derived `123_API_KEY` then failed at
the credential seam with a raw regular expression the user cannot act on.
The id must now start with a letter, and a test pins the relation between
the two rules rather than the regex.
A default naming a route the Models page has since removed left the
composer saying 选择模型 while the input still accepted a message, which
then failed inside the adapter mid-turn.
`session.prompt` now refuses with `model-unavailable` before opening a
turn. That is the enforcement boundary: the method stays callable no
matter what a client disables. `session.models` reports the same fact as
`routable`, and ui-model pushes a block through the new
`ctx.conversation.blocks` registry so the bar renders the disabled
textarea it already renders without a workspace, carrying the blocker's
own reason. The push direction is forced — ui-model already depends on
ui-conversation, so ui-conversation cannot read it back.
The gate is `routable`, not "matches no advertised group": catalog
membership is advisory, so a route serving a model it stopped advertising
is missing from the groups yet perfectly usable, and `null` before the
first load never blocks so a slow Host cannot lock a working composer.
The scaffold gains a route-only adapter for fixture-less keyless
scenarios. Registering zero providers is a test artifact — every product
composition mounts one — and the goldens that froze the seat's fallback
label now show the model those scenarios actually route to.
ChatView's bottom-follow recognized only wheel gestures as reader input,
so touch panning, scrollbar dragging, and keyboard paging could not leave
the bottom of a streaming transcript. Replace the wheel listener with
device-agnostic attribution: a scroll position deviating from the
observed-top ledger of the last delivered or written scrollTop is reader
input. Adds keyboard-paging and touch-style fling e2e scenarios (red under
the old implementation) and the bilingual Agent Note triplet.
ui-conversation now owns only the conversation.chat.turnTail hole; the row,
its derivation, and its copy live in @deepseek-ai/dsh-client-ui-deliverables,
composed in or out by one cordis.yml line.
The branch control on user and consumed-steering bubbles could enable
only when a completed turn ended with no node after the message at all,
so readers saw a permanently unavailable control whose tooltip promised
a state it could not reach. An enabled one would mislead anyway: a fork
at a message seq cuts at the containing turn/end and takes the answer
along, the opposite of the branch-to-re-ask reading a control on one's
own bubble suggests.
MessageItem loses its fork props, PendingSteeringBubble loses the
showBranch special case, and messageBranchSeqs narrows to
assistantBranchSeqs: only a completed turn's transcript tail that is
the turn's own content-text assistant may fork. A steered turn keeps
its fork point under the settled answer, because fork is a log-prefix
cut and the steer is model-visible history the child inherits.
Web aria goldens drop the user-bubble disabled-branch row and its
hidden explanation text; the nested-subagent golden also loses the one
enabled user-tail fork handle, a loss the decision note accepts.
Correct the Agent Note's consequence: a running turn withholds the footer
below its own trigger bubble, while every earlier completed turn keeps its
seat — which the package test already asserts.
Give the running-phase barrier an explicit budget: it is armed before the
park and awaited after the stop click, so the 30s replay default left no
headroom for the marker poll, the UI polls, and two aria captures.
Number the running-turn test's boundary seqs like the log does, with each
turn/end strictly after its own nodes.
`assistantActionsSeqs` picked the last content-text assistant of each turn
from the finalized transcript alone. That quantity is stable only once the
turn closes: while a turn is still producing steps, the narration written
before a tool call is the last content assistant so far, so copy, branch,
and the clock appeared under an intermediate sentence for as long as the
tool ran and then moved down to the next step's text.
Pass `ConversationSnapshot.turnEnds` into the derivation and grant the row
only inside a turn that has a durable `turn/end`. This is the same
completion fact the branch control and the `Ran for` label already read, so
the three parts of one row now agree; mid-turn narration owns nothing, and
the seat appears once under the settled answer.
`hasContentText` moves to chat-flow.ts so the ownership gate and
AssistantMarkdown's mount gate cannot drift apart.
apps/web/tests/turn-tail-actions.e2e.ts pins both states through the
assembled application: a hang sidecar on the second model call parks a turn
whose first step narrated before calling bash, and the two goldens hold the
parked flow and the flow after stopping.
ed30088adb deleted packages/ui/tui, so the two present-tense clauses naming
it in notes this branch already rewrites no longer describe HEAD. The
operative claim — the surfaces beyond the web row needed no change — stands
without naming the package. Pair hashes re-recorded.
Also anchor tool-row-styles.spec.ts's rule lookup at a rule boundary, so a
compound selector landing above a base rule fails loud instead of reading
the wrong declaration block.
`flex: none` stops the `+n` box shrinking but not its text wrapping, so a
row too narrow for title + separator + suffix broke the one-line summary in
the exact case the slot exists for. Add `white-space: nowrap` and pin the
declaration as CSS text, since jsdom has no layout.
Also align the README's row illustration with the English rendering (the
same sentence's plan-strip clause already used it), and record the two
deferred review findings in the Agent Note so they survive merge.
planSummary treated whitespace-only content as nameable, so a rejected call
whose args survive verbatim rendered a blank active clause beside a live +N.
The tool's own rule is trimmed non-empty; the row now uses it.
Also hoists the duplicated DSH_SNAPSHOT refresh flag out of the two assembled
snapshot files into their shared assembled-boot scaffolding.
The `+N` active count rides ToolRow's non-shrinking `summarySuffix` slot,
but only jsdom package suites covered it: the ACP snapshots render no web
surface and the built-boot smoke asserts no todo row. Add
`apps/web/tests/todo-row.snapshot.ts`, which boots the built client bundles
against the keyless fixture transport and records `summary`, `suffix`, and
the plan strip's header as separate fields, so folding the count back into
the summary string changes the expected output.
The three assembled-jsdom files now share `apps/web/tests/assembled-boot.ts`
instead of each carrying its own copy of the boot entry list, bundle map,
jsdom stubs, and mount call.
Also: name the policy branch in each `allowParallelInProgress` test title so
no case asserting `true` sits under a `false` describe, reword the stale cap
comment in todo-panel.spec.tsx, and record the plan strip's real header
format in the Agent Note (per-status counts, not `<done>/<total> tasks`).
Rebuild the todo row's parallel summary on the shared ToolRow that master
introduced: planSummary still returns the active name and the remaining
count separately, and the row hands the count to a new non-shrinking
summarySuffix slot so a narrow row clips the summary text before the count.
An error row drops the suffix, whose collapsed summary is the failure line.
Re-record the ACP todo-write transcript for the parallel prompt, regenerate
the config catalog for the required allowParallelInProgress field, and
re-record the bilingual pairing hashes.
- `relay` resolves its sender in `contextBody` like every other form. It was
the one shape whose marker could claim a form the body did not render: an
unreadable sender fell back inside the body while the row still said relay,
contradicting the contract this PR's own note states.
- `recall` requires the retained, omitted, and truncated fields. Completeness
is what the card exists to report, so a reference that cannot state it is
not a readable recall — showing the label alone presents a confident card
over unknown loss.
- The snapshot body states the supersession its producer framing line carries.
That line is the one part of the model-facing text no section contains, and
unlike an instruction context's `<system-reminder>` it states the form's own
semantics rather than wrapping content.
- `GoalMessageSource` is a discriminated pair, so `{ form: 'notice' }` without
its account no longer compiles. The guarantee this PR claims now holds at
that seam too, not only through `ContextFormed` on plugin sources.
- Goal and tool-goal summaries are bounded by a shared `boundContextSummary`,
which tool-tasks now uses as well. A goal objective is unbounded caller text
in exactly the way a task label is.
- The runtime snapshot interpolates once per request: agent-loop renders the
sections and joins them through `joinContextSections`.
- Every form's fallback branch is pinned, not only the notice one.
Four values complete the vocabulary, so the opaque body is reached only by
producers that genuinely promise no shape.
`snapshot` — current state a later snapshot supersedes. system-prompt now
exposes `renderContextSections()`, the named contributions
`renderContextSnapshot()` already joins for the model, so the body attributes
each part to the subsystem that produced it instead of re-splitting joined
prose. The runtime snapshot, time-context, and tmux-context declare it.
`notice` — a one-off account of what just happened, declared by tool-tasks,
goal state changes, tool-goal wrap-up, plan-mode switches, and
repeat-tool-guard. Its `summary` rides the COLLAPSED row: these five are the
majority of shipped producers and none of them needs expanding to be read.
The task summary bounds itself because its inputs are unbounded caller text.
`relay` — a message another agent addressed to this one; both subagent
sources declare it and the body names the sender above what it said.
`recall` — material lifted from another session's log. session-reference
needed no new field: its references already record retained and omitted
counts and the truncation flag, which the body shows first, because recalled
context is bounded on the way in.
`ContextFormed` is now discriminated by `form`, so a producer cannot declare
a shape without the facts that shape is presented from — a notice without its
summary, or a snapshot without its sections, fails to compile.
Only the two hook bridges stay opaque, by design: their content is whatever
an external program printed, so no shape can be promised for it. Unknown
kinds and unreadable records land there too.
- Content blocks render in the order the model received them. Partitioning
hoisted every unknown block past the text around it, so an interleaved log
read back in an order the model never saw.
- A delta distinguishes a newly reconciled file from a rewritten one; `set`
and `replace` already separate them at the producer, and collapsing both to
"updated" misread a new file.
- The superseded note states current fact in its consequences and testing
rather than keeping claims the implementation now contradicts, per
implemented/AGENTS.md; the decision itself stays as the record of that
change, with the supersession pointer above it.
The composer ring, percentage, and `~used / capacity` header read
`contextPressure.pressureTokens`, which moves only when a request reports
usage. Compaction reports none — compact-basic summarizes through a direct
`ctx.llm.stream()` call and appends only its own `compact/*` records plus the
replacement `user/message` — so the meter was frozen across the one action
taken to change it. Driving a real `compactNow` through the agent loop:
BEFORE compact: ring=4% header=~4227/100000 rows=[18, 0, 4365]
AFTER compact: ring=4% header=~4227/100000 rows=[18, 0, 286]
The composition rows fell 93%; the ring did not move, and would not until an
entire further turn completed. The panel then contradicted itself by more than
an order of magnitude at exactly the moment a reader opens it.
`contextPressure` now also publishes `projectedTokens`: the provider sample
plus the heuristic repricing of everything the surface gained or lost since
that sample, clamped at zero, folded through the shared `surface-fold.ts`. The
sample is stamped before the same event joins the surface, so an
`assistant/message` anchors against the surface its own request carried. Only
the delta is estimated, so the figure stays provider-anchored — the estimator's
CJK and JSON-schema underpricing stays out of the occupancy number — while
reacting the moment content lands or a span is shadowed. Same run after:
BEFORE compact: ring=4% header=~4323/100000 (pressure=4227, projected=4323)
AFTER compact: ring=0% header=~ 244/100000 (pressure=4227, projected= 244)
`contextOccupancy` prefers the projected figure and falls back to the bare
sample, so a projection restored from a pre-field checkpoint degrades to the
old behavior rather than disappearing. `stateVersion` moves to 3.
- An empty replacement catalog is a real catalog: `renderCatalogUpdate()`
publishes zero entries when the last skill disappears, and falling back
would hide that every earlier name was retired.
- The opaque fallback keeps a `form` declaration this version cannot present.
It is the one place a newer or foreign log's declared shape would otherwise
vanish from the UI entirely, since the row marker is also absent there.
- An instruction change with an unrecognized `action` disqualifies the record.
The action decides the word the row shows, so an unknown one would be
presented as loaded or updated.
- The catalog list bounds itself and reports the withheld count. Entry count
is unbounded and the scrollport bounds height, not node count.
- A catalog message keeps content blocks this version does not know, instead
of dropping model-visible content the extensible union may carry.
- `core.md` defines `ContextFormed`, the interface actually carrying the
optional field, beside `ContextForm`.
- The superseded-in-part bullet states the affected clauses as one rule rather
than enumerating them; two rounds of enumeration each missed some, which is
the shape being fragile rather than the list being wrong.
- The note records the one migration case that does not self-heal: an
old-format catalog as the only one, with an empty current view, leaves a
stale catalog nothing replaces.
The panel header concatenated a `45%` span with a `context.used` fragment, so
Chinese rendered "45% 上下文已用" against the ring's own "上下文已用 45%". The
header now renders the one localized `context.aria` sentence split around its
`{percent}` slot: each locale owns the reading's position while the reading
keeps its primary tone, and the side a locale leaves empty collapses through
`.headline:empty` instead of spending a header gap.
The bar mapped every composition row to a segment unconditionally, and
`.segment`'s 2px min-width kept each one visible, so a 0% occupancy panel
painted an ~8px filled bar over an empty context. Parts are now computed with
their widths and zero-width parts are filtered out, which also collapses the
plain and segmented branches into one map.
- `catalogHistory` validated its durable read. `agent.session.events` is a
JSONL/SQLite seed on resume or fork, and seed validation guarantees only a
source object with a non-empty `kind`; a `skill-catalog` record with missing
or wrongly shaped `entries` threw inside the step listener, failing every
later turn of that session. It is now skipped as an unrecognizable record,
the posture the replaced content digest had, with a regression test over six
malformed shapes.
- The headless keyless smoke still filtered catalogs by the old plugin source,
so the `built-bin-smoke` gate would not have found the catalog message.
- Entries record the published description unescaped. The pseudo-XML escaping
belongs to the `<available_skills>` frame and is applied when rendering it,
so a description containing `<` no longer reaches the card as `<`.
`escapeText` is injective, so republish semantics and the model-facing text
are unchanged.
- Adjacent text blocks join with no separator, matching how provider adapters
flatten them; the body no longer shows a line break the model never saw.
- Provenance fields are bounded like the text: an unknown producer may record
an arbitrarily large value.
- Both readers are all-or-nothing, and the row's form marker reports what
rendered rather than what was declared, so a partly unreadable record cannot
present a confident but incomplete account.
- The catalog body consumes `update` as a replacement notice; the digest
canonicalizes per entry as JSON, since every separator character is itself
legal in a description.
- `core.md` documents the form axis with a `ContextForm` type-equiv block, and
both projections assert the wiring they duplicate.
Assistant footers and the stats line gain TTFT/tok-per-second readings
folded from step timings; context occupancy moves off the stats line onto
a composer ring whose panel shows a heuristic system/tools/messages
breakdown from the new token-meter contextBreakdown session projection.
Injected context reached the transcript as one anonymous shape whose
expanded body was the whole message serialized as inline JSON, so the
escaping collapsed the only readable part — the model-facing prose —
into a single line.
`MessageSource` gains an optional producer-declared `form`, a small
semantic vocabulary of information shapes independent of `kind`: kind
says who produced the context, form says what shape it is, so several
producers may share one presentation. Two values ship.
`instructions` (workspace-context) lists the reconciled files above
their text and keeps the `<system-reminder>` framing verbatim, because
the framing is part of what the model read. `catalog` moves dsh-tool-skill
off the shared plugin kind onto a `skill-catalog` source carrying the
published name/description entries, and the body lists those instead of
re-parsing `<available_skills>` out of the prose. Catalog identity moves
with it: the republish digest now covers the durable entries, deleting
the text-slicing that recovered them from a logged message.
Everything else renders the opaque body — the model-facing text with its
real line breaks, then the remaining provenance as fields. That is the
documented default, not a leftover: a resumed, forked, or foreign log
must render whether or not its producer is mounted here, which is why
the classification lives in the durable source rather than a client-side
table keyed by producer.
A completed pwsh foreground call now presents as the bash tool's terminal
card with the parsed exit-status pill instead of a generic console fence,
and the collapsed row classifies as the shell family (Pwsh title). The
marker-to-exit-status parse moves from dsh-tool-bash's private render
module into the @deepseek-ai/dsh-bash seam so both shell tools share one
inverse of the marker contract (the bash-env precedent). The Web UI needs
no per-tool code: the terminal-card bridge maps any card:'terminal' view.
Coverage: dsh-bash owns the parse edge cases; the tool-pwsh presenter
suite mirrors tool-bash's; the client row-model suite pins the Pwsh shell
row; the new keyless pwsh-terminal web lane seeds an authored session,
presents it through the real tool on replay, and pins the card golden.
TUI is out of scope: the TUI package was removed (ed30088adb), so the Web
surface is the only UI the gap affected; the roadmap proposal's stage 2
is updated accordingly.
Yichen Jiang
imccyu
fz
creatixchu
Chinesezjc
ZiyaZhang
imccyu
Ziya
Huanqi Cao
Yif