- Mount WorkspacesService.startInitialSelection in the runtime apply (the
one-shot baseline follower shipped in 98633b5aa without a caller): a
restored current session wins, an explicit clear stays cleared, a failed
connect retries on the next baseline projection.
- Cover the policy in client-apply and the assembled workspace-flow
snapshot; startup now lands in the recent Workspace's blank session, so
the draft-carry scenario starts from the hero directly.
- Bring docs along: startup-selection paragraphs in the session-scope RFC
note (both languages), bilingual README pairs for the four new client
packages, doc-graph regeneration with client-declared events exempt from
the dispatcher requirement (client dispatch sites are structurally
invisible to the host-side ts.Program), and pairing re-records.
- Keep ConversationSession mounted for blank sessions (chrome-less) so the
draft-persistence mirror stays bound in the hero; hero typing reaches the
chat store again.
- Restore the baselines-ready gate in AppFrame: empty boot snapshots no
longer flash the New Workspace hero before either baseline lands.
- Commit ordinary sends through the machine (send-committed event +
Shell.commitSend): undo can no longer resurrect already-sent content on
the default-sink path.
- Give the production InputMachine a real wall clock so the typing-run
merge window actually expires.
- Coalesce concurrent connectWorkspace creates per workspace: the summary
has no cwd until the host frame lands, so a second New Session inside
that window minted a duplicate hidden blank session.
Same client-lane debt as the existing GUI exclusions (TODO(gui)): the new
ui-slash/ui-command/ui-skill/ui-sidebar/ui-workspace client files and the
connection fixture keep their uncovered branches until the browser-grade
harness lands.
Test-side catch-up with the session-maybe conversation architecture: the
provide channel's descriptor shape and maybeProvideInfo in fakes, the shared
chat-store handle asserted on conversation.session (the session-maybe shell
carries no store), startSession fakes exposing the workspace list snapshot,
strict session slots declining (not throwing) without a session, AppFrame's
removed empty seat and loading gate, and the hero draft asserted on the
machine (the chat-store mirror binds with ConversationSession). Plus three
lint fixes (max-len split, boolean-compare, arrow-parens/unbound-method).
eslint --fix autofixes plus manual repairs: max-len line splits
(fake-api handlers, notifier/slots JSDoc, spec signatures), charAt over
non-null-asserted indexing in slash detect/menu cores, Array.from for
code-point capping, typeof assertions for unbound-method in specs,
generic getByRole for the send-button cast, effect disposer void-wrap in
command register, and dropped unused type imports.
Every PR that changes product-user-visible GUI behavior now includes a
demonstration GIF with real provenance (that branch's built tree, real
key, real model rounds). Recording stays side-effect-free; the skill
gains a bounded final publication step: GIFs go on an append-only
orphan assets branch (one per PR series) and embed via the blob URL
with ?raw=true, never on the PR branch itself.
Folds in the operational lessons from the Code Mode UI series:
.playwright-mcp/ screenshot roots (now gitignored), per-PR staging and
precise server teardown, one-call DOM polling for transient states,
exact-text completion predicates, prompt engineering for UI states,
and the export-before-invoke GIF_SKILL_DIR encoder pitfall.
Agent Note: implemented/process/2026-07-26-gui-pr-gif-evidence-and-assets-branch (+ zh pair); the 2026-07-23 recording note now defers publication policy to it.
Responding to review on #661: a public method on the generic ToolRegistry
service whose only caller is the run_code bridge was ad-hoc surface
widening. The bridge now receives it as a registry-private capability
closure in RunCodeBridgeOptions (the requireRuntime idiom, alongside the
cap), the method is private, and it leaves the generated service
catalog/API surfaces.
The pattern is now named as a code smell where reviewers look: the
packages/AGENTS.md capability-interface rule gains the inverse-smell
clause (ceiling 660→675 — the list is at capacity and the clause needs
one sentence), and dsh-code-review's capability-fit check tells reviewers
to flag single-consumer public service methods and require the closure
form.
Validate replay sidecars and cross-copy failure facts, make browser console tripwires and macOS temp paths deterministic, and wait for asynchronous TUI resume details. Keep the owning docs, translations, and generated catalog aligned.
