🤖 refactor: auto-cleanup

[mux-bot]

Summary

Rolling, low-risk cleanup batch. Each run adds one extremely small,
behavior-preserving refactor picked from changes that have landed on main
since the last checkpoint.
Auto-cleanup checkpoint: 789053a

Current diff against main:

• src/browser/utils/additionalSystemContextStore.ts — drop the
  unused __resetAdditionalSystemContextFocusForTests export and
  promote pendingFocusGeneration from let to const. The reset
  helper was added speculatively alongside the Instructions tab in
  🤖 feat: Instructions tab in right sidebar #3262, but ripgrep across src/ and tests/ finds zero importers —
  no test ever calls it. Removing it lets the only other reason the
  module needed a mutable binding (the pendingFocusGeneration = new Map() reassignment inside the helper) go away, so the binding can
  be a const Map that's still mutated in place via .set(). The
  requestAdditionalSystemContextFocus /
  getAdditionalSystemContextFocusGeneration callers continue to see
  the exact same Map identity and counter semantics — pure dead-code
  removal with no runtime behavior change.

• src/browser/features/Messages/Mermaid.tsx — hoist the URL-attribute
  set and blocked-scheme list out of sanitizeMermaidSvg into
  module-level constants (SANITIZER_URL_ATTRIBUTES,
  SANITIZER_BLOCKED_URL_SCHEMES). They are pure lookup data with no
  per-call dependency, so allocating them on every Mermaid render was
  wasted work; they now allocate once per module load. Identical
  contents and identical lookup semantics — pure cleanup with no
  behavior change.

• src/node/services/agentStatusService.ts — hoist the AI SDK v5
  tool-part state → lifecycle phase mapping out of summarizeToolPart
  into a module-level TOOL_PART_PHASE_BY_STATE record. The previous
  nested ternary obscured the actual mapping (output-available and
  output-redacted both fold to "done") and made adding new states a
  multi-line edit. The summarizer now does a single table lookup; the
  union-typed values document the two valid phases inline. States not
  in the table (e.g. input-streaming) still yield null and the bare
  [tool <name>] form, byte-for-byte identical to before.

• src/browser/utils/slashCommands/experimentVisibility.ts (+ two
  consumers) — add createSlashCommandExperimentResolver(snapshot)
  alongside resolveSlashCommandExperimentValue. Three slash-command
  discovery surfaces (suggestions in ChatInput, the ghost-hint call in
  ChatInput, and the CommandPalette palette query) had each inlined
  the same (experimentId) => resolveSlashCommandExperimentValue(experimentId, snapshot) lambda. The new helper returns the exact same closure each
  callsite previously built inline, so the isExperimentEnabled
  predicate handed to getSlashCommandSuggestions / getCommandGhostHint
  is behaviorally identical; only the wiring detail (that resolution
  is parameterized by experiment id) moves into the visibility module.
  Rebased through the goals GA removal — the snapshot now carries
  only workspaceHeartbeats but the resolver helper itself is
  unchanged.

• src/cli/run.ts — fold the Runtime type import into the existing
  import type { InitLogger, WorkspaceInitResult } from "../node/runtime/Runtime"
  line. The SSH-provisioning commit (🤖 perf: cut SSH workspace startup ~9× on the warm path via fused single-RTT materialization #3302) added a second import type { Runtime } from "../node/runtime/Runtime" directly below it; merging
  them is byte-identical semantics with one fewer line of boilerplate.
  No behavior change.

• src/node/services/agentSkills/agentSkillsService.test.ts — extract a
  shared BUILT_IN_SKILL_NAMES constant for the two toEqual([...])
  assertions that previously enumerated the same five built-in skill
  names (imagegen, init, mux-diagram, mux-docs, orchestrate)
  inline. The two call sites differ only in their project-skill prefix;
  spreading the shared constant at the end of each list preserves the
  exact sort order and assertion semantics. The next built-in skill
  added (e.g. when /orchestrate graduates to advertised, or a new
  hidden skill lands) only needs to update one list instead of two.
  Byte-equivalent assertion output; no runtime behavior change.

• src/browser/features/Settings/Sections/TasksSection.tsx — extract an
  AdvisorSwitch component for the per-agent advisor toggle. The
  Tooltip + Switch + optional Reset button block was duplicated
  between renderAgentDefaults and renderUnknownAgentDefaults (added
  in 🤖 fix: enable advisor for default subagents #3309 when the Advisor experiment gained default-on behavior for
  Exec and Plan). The two call sites differ only in the agent-id
  variable they close over and which setter/resetter to invoke, so the
  shared markup now lives in one place. Same aria-label, same
  tooltip text via getAdvisorSwitchState, same conditional Reset
  semantics — the TasksSection UI test (defaults advisor on for Exec and Plan when the experiment is enabled) still finds both switches
  by their unchanged role + name.

• src/common/utils/ai/streamChunks.ts (+ two consumers) — hoist the
  ["text", "delta", "textDelta"] field priority used to extract AI SDK
  v5 text-delta payloads into a shared TEXT_OUTPUT_DELTA_FIELDS
  constant. Both streamManager.ts (the fullStream text-delta part
  handler) and tools/advisor.ts (the advisor streamText onChunk
  text-delta handler, added in 🤖 feat: stream advisor output live #3310) had inlined the same array
  literal; centralizing it prevents drift if the AI SDK normalizes
  another alias and documents the intent in one place. The other two
  extractChunkDeltaText callsites in aiService.ts use deliberately
  different field orderings (["textDelta", "delta", "text"] for
  parent-step text-delta capture, ["text", "textDelta", "delta"] for
  reasoning-delta capture) and are intentionally left untouched.
  Byte-equivalent lookup semantics at every callsite — pure DRY.

• src/browser/features/RightSidebar/GoalTab.tsx — extract a small
  file-scoped StatTileHeader component for the duplicated <dt> +
  inline-Edit-button header row used by BudgetTile and TurnsTile
  (added in 🤖 feat(goals-tab): consolidate Cost/Budget/Remaining + always show Turn cap #3314 when the active-goal stat surface was consolidated).
  The two tiles previously inlined the same flex wrapper, the same
  text-muted text-xs <dt>, and the same text-xs underline Edit
  button — only the label string and aria-label differed. The shared
  helper preserves those two parameters as props and keeps every class
  name, type="button", and aria label byte-for-byte identical, so the
  inline budget/turn-cap editors in GoalTab.test.tsx (which locate
  their openers via getByLabelText("Edit goal budget") /
  getByLabelText("Edit goal turn cap")) still match. Pure DRY; no
  behavior change.

• src/common/orpc/types.ts (+ three consumers) — hoist the
  GoalInterventionPolicy = NonNullable<SendMessageOptions["goalInterventionPolicy"]>
  alias next to SendMessageOptions itself. After 🤖 feat: add goal intervention policy #3319 added the
  policy field, three independent files re-derived the same alias:
  the browser ChatInput/types.ts barrel, node/services/messageQueue.ts,
  and node/services/agentSession.ts. The canonical alias now lives
  beside its source schema; ChatInput/types.ts re-exports it so the
  feature-local import surface stays unchanged, and the two node-side
  files import directly. Identical type semantics at every callsite —
  pure relocation with no runtime behavior change.

• src/node/services/agentSession.ts — convert
  synthesizeSilentContinuationSummary from a private method on
  AgentSession to a module-level free function. The helper, added in
  🤖 fix(goals): treat text-only continuation turns as goal completion #3326 as part of the silent-continuation auto-complete hook, only
  depends on its input parts argument — it never touches this.
  Moving it alongside the existing free helpers
  (stripGoalInterventionPolicy, coerceGoalSyntheticMessageKind,
  extractAgentSkillRefs, etc.) makes the no-this contract obvious
  to readers and keeps the file's helper style consistent. The
  companion maybeAutoCompleteGoalFromSilentContinuation method stays
  on the class because it accesses this.workspaceGoalService,
  this.workspaceId, and the instance-scoped log. Byte-for-byte
  identical logic at the only callsite; the
  agentSession.goalAutoPause.test.ts suite (18/18 pass) continues to
  cover the silent-continuation, dynamic-tool-present, manual-user, and
  paused-goal branches.

• src/node/services/tools/review_pane.ts — switch the dedup seed loop
  in applyReviewPaneUpdate from for (const h of base) { seen.set(..., base.indexOf(h)) } to for (const [i, h] of base.entries()) { seen.set(..., i) }. The previous form ran an O(n²) reference scan on every seed
  iteration even though the loop already had the index in hand; the
  result is byte-equivalent for the realistic case where every entry in
  base is a unique reference (which is always true today — base is
  either [] or [...current] after the previous deduped call) and now
  correct-by-construction even if a future refactor ever introduced
  duplicate references. Same dedup map contents, same final hunk
  ordering — the 15 existing review_pane.test.ts tests (including the
  "dedupes by formatted path:range key when adding, preferring the
  latest comment" case) all pass unchanged. Picked from the
  review_pane_update tool feature landed in 🤖 feat: add review_pane_update tool + Assisted review toggle #3331.

• src/cli/run.ts — extract a CLI_DISABLED_TOOL_NAMES tuple and build
  the mux run toolPolicy array via .map(name => ({ regex_match: name, action: "disable" as const })). The five entries previously inlined
  in buildSendOptions (ask_user_question, status_set, todo_write,
  todo_read, notify) differed only by tool name and all shared the
  exact same action: "disable" shape. Picked from the
  ask_user_question disable that landed in 🤖 fix: disable ask_user_question in mux run #3336 — that commit added
  a fifth identical-shape entry, so the table form is now strictly
  tidier than five copy-pasted object literals, and the next disabled
  CLI tool drops to one string instead of duplicating the object
  shape. Byte-equivalent toolPolicy contents (same five entries,
  same order, same regex_match and action values), so no behavior
  change.

• src/browser/features/Tools/AskUserQuestionToolCall.tsx — extract a
  module-level getSectionPillClassName(isSelected, isComplete) helper
  for the executing-UI section pills. The two pill renderers (one per
  question, plus the Summary pill, both added in 🤖 refactor: polish ask_user_question UI #3339 when the
  ask_user_question UI was polished) inlined the exact same nine-class
  Tailwind string and the same three-way selected (accent) / complete
  (success-tinted) / pending (neutral) branch logic — they differed
  only in which boolean drove the "selected" leg (isActive for
  question pills, isOnSummary for the Summary pill) and which boolean
  drove the "complete" leg (answered vs isComplete). The helper
  centralizes the class string so the next visual tweak to those pills
  lives in one place. Byte-equivalent class output at both callsites
  (verified by diffing the rendered class strings), so the existing
  AskUserQuestionToolCall.test.tsx suite (4/4 pass) continues to
  locate the buttons by their unchanged role + accessible name.

• src/node/services/signingService.test.ts — extract a module-local
  restoreEnvVar(key, savedValue) helper to fold the four inlined
  copies of the SSH env-var restore block (twice in the
  withoutSshAgent finally, twice in afterAll) into single-line
  calls. Each copy used the identical if (saved === undefined) delete process.env[X]; else process.env[X] = saved shape — only the env-var
  name differed (SSH_AUTH_SOCK vs SSH_AGENT_PID). The helper
  preserves that exact delete-vs-assign decision per key, so the
  process.env state after withoutSshAgent returns and after the suite
  tears down is byte-equivalent to before. Picked from 🤖 tests: stabilize flaky signing and best-of task tests #3340, which
  brought this file back under review when it bumped the Ed25519
  "should load key and return capabilities" test to a 15s timeout. The
  full suite (14/14 SigningService tests, including all withoutSshAgent
  paths and the agent-vs-disk priority tests that depend on the
  restore semantics) passes unchanged. Pure DRY; no behavior change.

• src/browser/features/RightSidebar/CodeReview/ReviewPanel.tsx
  (+ test consumer) — relocate the test-only buildReviewDiffPathFilter
  single-result wrapper out of ReviewPanel.tsx and into
  ReviewPanel.assistedStats.test.ts as a private file-scoped helper.
  The wrapper was introduced in 🤖 fix: surface assisted review hunks reliably #3344 alongside the multi-spec
  buildReviewDiffPathFilterSpecs form so the existing
  buildReviewDiffPathFilter describe block (two assertions: assisted
  mode fetches agent-pinned files, non-assisted mode preserves the
  selected file pathspec) could keep using a pathFilter: string
  return shape. The two production callers in the same file
  (ReviewAssistedStatsReporter's diff-load effect and
  ReviewPanel's diff-load effect) already use the multi-spec
  buildReviewDiffPathFilterSpecs directly so they can fan out across
  per-repo-root pathspecs — nothing in production ever called the
  single-result wrapper. The relocated helper has byte-identical
  implementation (same ...params, projectPath: repoRootProjectPath ?? "" forwarding, same [0]?.pathFilter ?? "" first-spec
  extraction), so the 10/10 ReviewPanel.assistedStats.test.ts
  assertions pass unchanged; ReviewPanel.tsx just stops exporting a
  helper that no production code consumes.

• src/cli/run.ts — convert the two throw new Error('Invalid --goal-* "' + value + '" ...') messages in the new parseGoalBudgetFlag and
  parseGoalTurnsFlag helpers (added in 🤖 bench: add Terminal-Bench strict goal mode #3348) to backtick template
  literals (`Invalid --goal-budget "${value}". ...`). These were
  the only two throw new Error('...') single-quoted-concatenation
  pairs left in src/cli/run.ts; every other throw new Error( in the
  file — including the adjacent parseMode helper at the top of the
  same block — already uses a backtick template literal. The thrown
  Error message text is byte-for-byte identical (same Invalid --goal-budget "<value>". Expected dollars like 5, $5.00, or cents like 500c and Invalid --goal-turns "<value>". Expected a positive integer strings, same single quotes around the user value), so this
  is a pure stylistic alignment with no behavior change.

• src/browser/features/RightSidebar/GoalTab.tsx — hoist the duplicated
  neutral-button Tailwind class string used by the Pause / Mark complete
  / Reopen lifecycle action buttons into a single module-level
  GOAL_LIFECYCLE_NEUTRAL_BUTTON_CLASS constant. After 🤖 refactor: style goal row actions as real buttons #3342 reshaped
  the completed-goal action row, three buttons in the same flex flex-wrap gap-2 group inlined the same nine-class string
  (border-border-light bg-surface-secondary text-foreground hover:bg-surface-tertiary inline-flex items-center gap-1.5 rounded-md border px-3 py-1.5 text-sm); the accent-colored Archive and
  success-tinted Resume buttons keep their own inline strings because
  they are the only primary / positive variants in the row. The
  existing active goals expose a de-prominent Clear; completed goals promote Archive as primary assertion (reopenButton.className does
  NOT contain bg-accent; archiveButton.className DOES) and every
  Pause / Mark complete / Reopen aria-label lookup in
  GoalTab.test.tsx (23/23 pass) continue to hold. Byte-equivalent
  class string at every callsite; no behavior change.

• src/node/runtime/SSHRuntime.ts — dedupe the four receive-pack
  thin-pack failure patterns (unresolved deltas, unpacker error,
  unpack-objects abnormal exit, remote unpack failed) that 🤖 fix(ssh): harden bare base repo against receive-pack thin-pack failures #3356
  introduced inline in both PROJECT_SYNC_RETRYABLE_ERRORS (used by
  the broad retry classifier at the isProjectSyncRetryableError
  callsite) and UNRESOLVED_DELTA_PUSH_PATTERNS (used by
  isUnresolvedDeltaPushFailure to decide whether the next retry
  attempt should add --no-thin). The sub-class is now declared once
  on UNRESOLVED_DELTA_PUSH_PATTERNS (with the long rationale
  consolidated onto its JSDoc) and spread into
  PROJECT_SYNC_RETRYABLE_ERRORS via ...UNRESOLVED_DELTA_PUSH_PATTERNS.
  Both .some((pattern) => errorMsg.includes(pattern)) iterations see
  the same four patterns in the same order, so retry classification and
  the forceNoThinNextAttempt decision are byte-equivalent — pure DRY,
  no behavior change. The existing SSHRuntime.retry.test.ts and
  SSHRuntime.syncContract.test.ts suites both pass unchanged.

• src/browser/features/Tools/bashCollapsedSummary.ts — extract a
  matchesCommand(value, command) helper next to normalizeForComparison
  and route the three existing normalizeForComparison(X) !== normalizeForComparison(command) callsites through it. The intent-only
  summary mode added in 🤖 feat: add intent-only bash summaries #3360 introduced a second
  normalizeForComparison(displayName) !== normalizeForComparison(command)
  inside the new getIntentOnlyFallback, doubling the predicate that was
  already used to filter a redundant intent and inside
  stripTrailingUsingCommand to detect a redundant trailing using <command> clause. The helper composes the existing normalizer, returns
  the identical boolean, and no callsite changes its branch direction —
  so intent-vs-command redundancy filtering, display-name fallback
  selection, and the using-clause stripper are all byte-for-byte
  identical. The existing bashCollapsedSummary.test.ts suite (10/10)
  exercises every branch — intent equal to command, intent missing,
  intent-only mode falling back to display name, display name equal to
  command, and the nested using-clause sanitize pass — and all pass
  unchanged.

Validation

• make static-check — eslint, tsgo (×2 configs), and prettier all green
  locally; the only red step is fmt-shell-check failing because shfmt
  isn't installed in this environment (pre-existing env limitation, no
  shell files touched by these changes).
• bun test src/browser/features/RightSidebar/GoalTab.test.tsx — 23/23 pass.
• bun test src/browser/features/RightSidebar/CodeReview/ReviewPanel.assistedStats.test.ts
  — 10/10 pass.
• bun test src/browser/features/Tools/AskUserQuestionToolCall.test.tsx
  — 4/4 pass.
• bun test src/node/services/signingService.test.ts — 14/14 pass.
• bun test src/browser/features/Tools/bashCollapsedSummary.test.ts — 10/10 pass.

---

Generated with mux • Model: anthropic:claude-opus-4-7 • Thinking: xhigh

[mux-bot]

⚠️ Auto-fixup could not resolve CI failure: Windows-only flake in tests/ipc/runtime/backgroundBashDirect.test.ts > Foreground to Background Migration > should migrate foreground bash to background and continue running.

Failure:

expect(received).toContain(expected) // indexOf
Expected substring: "BEFORE_BG_fg_to_bg_1779223264672"
Received string:    "Process sent to background with ID: fg_to_bg_1779223264672
Output so far (0 lines):
"
    at tests/ipc/runtime/backgroundBashDirect.test.ts:410:27

Why this is a flake, not caused by the cleanup commit:

• This test is already known to be timing-sensitive on Windows. The file defines FOREGROUND_MIGRATION_READY_MS = process.platform === "win32" ? 900 : 300 and there's an existing commit titled "Harden fast-exit foreground migration assertion on Windows" targeting this exact file.
• The "Output so far (0 lines)" shape means the foreground bash hadn't flushed its first echo line before sendToBackground was called — a Git Bash startup / IO timing race on Windows CI.
• The auto-cleanup diff touches none of: background process manager, LocalRuntime, bash tool, runtime/, or this test. It only touches UI components, agent session/status/queue/stream services, image/advisor/review_pane tools, agentSkills test, experimentVisibility, cli/run, orpc/types, and streamChunks.

Action: No fix pushed. Recommend retrying the Windows job; if it persists, the Windows ready-wait may need to be bumped above 900ms or the assertion further hardened (separate from this cleanup PR).

[mux-bot]

⚠️ Auto-fixup could not resolve CI failure: pre-existing process-wide test pollution flake, unrelated to this PR.

Symptom (run 26251744959): 78 Test / Unit failures cascading from TypeError: window.addEventListener is not a function originating in src/browser/utils/RefreshController.ts:455, with downstream collateral damage in GitStatusStore, WorkspaceHeartbeatModal, InlineSkillMarkdown, goal set objective prompt, task_apply_git_patch, etc.

Root cause: src/browser/stores/WorkspaceStore.test.ts:148 assigns global.window = mockWindow (a partial stub lacking addEventListener). Bun runs all test files in the same process, so any later file that constructs a RefreshController crashes on the leaked partial window.

Why this isn't the cleanup PR's fault:

• This PR's diff touches 26 files of pure refactors (constants extraction, helper extraction, type relocation). None of them modify RefreshController.ts, GitStatusStore.test.ts, WorkspaceStore.test.ts, usePersistedState.ts, tests/setup.ts, or any window/global-handling code.
• Running the suspected pair locally (bun test src/browser/stores/WorkspaceStore.test.ts src/browser/stores/GitStatusStore.test.ts) passes 150/150 — the failures only manifest when the full bun test src suite runs in process-wide order.

Existing fixes on other branches (not yet on main):

• e477ac3e7 on assisted-review-ux-pass — fix(tests): guard window listeners against partial-window test stubs (adds typeof X === "function" guards in RefreshController constructor/dispose and usePersistedState). Its commit message explicitly notes this affects "unrelated test files even though each passes in isolation".
• 95312509c on benchmark-m445 — tests: stabilize GitStatusStore window mock.

Recommendation: cherry-pick e477ac3e7 (or fix WorkspaceStore.test.ts to mock a complete window) onto main as a standalone fix; once merged, re-run CI here. Manual intervention needed.