Merge origin/master into worktree/fix-question-drafts-session-switch

apps/web/tests/agent-preset-selection.e2e.ts

@@ -9,7 +9,8 @@
 //
 // Zero model calls: no replay fixture mounts, so a stray stream fails loud.
 import { fileURLToPath } from 'node:url'
-import { mkdir, writeFile } from 'node:fs/promises'
+import { mkdir, mkdtemp, realpath, writeFile } from 'node:fs/promises'
+import { tmpdir } from 'node:os'
 import { join } from 'node:path'
 import type { Browser, Page } from 'playwright'
 import { chromium } from 'playwright'
@@ -22,7 +23,9 @@ import {
   captureStableAria, compareOrRefreshGolden, launchWebScaffold, seedSession, watchConsole,
   webSnapshotMode, type WebScaffold,
 } from './scaffold.ts'
-import { connectFreshWorkspace, newEnglishPage, saveFailureShot } from './support.ts'
+import {
+  connectFreshWorkspace, newEnglishPage, saveFailureShot, writeComposerDraft,
+} from './support.ts'
 
 const SNAPSHOT_DIR = fileURLToPath(new URL('./expected/agent-preset-selection', import.meta.url))
 const HERO_EXPECTED = join(SNAPSHOT_DIR, 'hero.expected.md')
@@ -32,6 +35,26 @@ const MODE = webSnapshotMode()
 const SEED_ID = 'agent-preset-selection-web-e2e'
 /** A project skill only a preset that mounts `skill-filesystem` can discover. */
 const SKILL_NAME = 'preset-catalog-demo'
+/** The preset whose rows resolve and then refuse to start. */
+const REFUSING_ID = 'zz-refusing'
+
+/**
+ * Seed a preset discovery reports healthy and the mount refuses.
+ *
+ * Every row resolves — the module is right there beside the composition — so
+ * health has nothing to report and the chip offers the preset like any other.
+ * Only starting it finds out, which is the case the chip's banner exists for.
+ * @param root - the lane's writable preset root.
+ */
+async function seedRefusingPreset(root: string): Promise<void> {
+  const directory = join(root, REFUSING_ID)
+  await mkdir(directory, { recursive: true })
+  await writeFile(join(directory, 'refuses.mjs'),
+    'export const name = \'refuses\'\nexport function apply() { throw new Error(\'this row refuses to start\') }\n')
+  await writeFile(join(directory, 'agent.cordis.yml'), '- id: refuses\n  name: ./refuses.mjs\n')
+  await writeFile(join(directory, 'preset.yml'),
+    'name: Refusing mode\ndescription: Resolves, then refuses to start.\n')
+}
 
 /**
  * Seed one project skill under the connected workspace.
@@ -176,11 +199,16 @@ describe('web e2e: agent-preset selection', () => {
   let browser: Browser
   let page: Page
   let tripwire: ReturnType<typeof watchConsole>
+  let presetRoot: string
 
   beforeAll(async () => {
-    // The scaffold's default roster pin is exactly this scenario's shape: the
-    // plugin's shipped presets, default `standard`.
-    scaffold = await launchWebScaffold({})
+    // The shipped presets, plus one lane-owned preset that mounts and refuses:
+    // the chip's own failure path needs a preset the roster offers.
+    presetRoot = await realpath(await mkdtemp(join(tmpdir(), 'dsh-web-e2e-refusing-')))
+    await seedRefusingPreset(presetRoot)
+    scaffold = await launchWebScaffold({
+      agentPresets: { roots: [{ path: presetRoot, trust: 'user' }], default: 'standard' },
+    })
     // A resumed session runs what it was created with; seeding one that
     // records `minimal` is what makes the header label a claim about the
     // session rather than an echo of the current default.
@@ -237,8 +265,23 @@ describe('web e2e: agent-preset selection', () => {
     await expect.poll(() => livePreset(scaffold), { timeout: 15_000 }).toBe('minimal')
   })
 
+  it('says why a switch was refused instead of letting the chip revert in silence', async () => {
+    onTestFailed(() => saveFailureShot(page, 'web-e2e-agent-preset-refused'))
+    await page.getByRole('button', { name: 'Minimal mode' }).click()
+    await page.getByRole('menuitem', { name: /Refusing mode/ }).click()
+
+    // Health cleared every row, so nothing on the settings page says this
+    // preset is unusable — the banner is where the host's reason lands, and
+    // without it the chip just snaps back to the preset it already ran.
+    const banner = page.getByRole('alert').filter({ hasText: 'Refusing mode' })
+    await banner.waitFor({ timeout: 15_000 })
+    expect(await banner.textContent()).toContain('this row refuses to start')
+    await expect.poll(() => livePreset(scaffold), { timeout: 15_000 }).toBe('minimal')
+    await page.getByRole('button', { name: 'Minimal mode' }).waitFor({ timeout: 10_000 })
+  }, 60_000)
+
   it('re-reads the slash catalog through the composition the switch installed', async () => {
-    // Continues the previous case: the chip has already applied `minimal` to
+    // Continues 'applies the staged pick': the chip has already applied `minimal` to
     // the blank session, and this one reads the menu that switch left behind.
     onTestFailed(() => saveFailureShot(page, 'web-e2e-agent-preset-slash-catalog'))
     const composer = page.locator('[data-composer-input][contenteditable="true"]').last()
@@ -246,7 +289,7 @@ describe('web e2e: agent-preset selection', () => {
     // `minimal` mounts neither the compaction group nor plan mode nor local
     // skill discovery, so the catalog the composer warmed under the
     // deployment default must not survive the switch.
-    await composer.fill('/')
+    await writeComposerDraft(page, composer, '/')
     await expect.poll(() => menuOptions(page), { timeout: 15_000 })
       .not.toEqual(expect.arrayContaining([expect.stringContaining(SKILL_NAME)]))
     const onMinimal = await menuOptions(page)
@@ -256,7 +299,7 @@ describe('web e2e: agent-preset selection', () => {
     // remains outside every preset.
     expect(onMinimal.some(option => option.startsWith('goal'))).toBe(false)
     expect(onMinimal.some(option => option.startsWith('model'))).toBe(true)
-    await composer.fill('')
+    await writeComposerDraft(page, composer, '')
 
     // Switching back up reaches the host at all — the chip compares the pick
     // against its list row, so a row that never reprojected the first switch
@@ -266,14 +309,14 @@ describe('web e2e: agent-preset selection', () => {
     await page.getByRole('menuitem', { name: /^Standard mode/ }).first().click()
     await expect.poll(() => livePreset(scaffold), { timeout: 15_000 }).toBe('standard')
 
-    await composer.fill('/')
+    await writeComposerDraft(page, composer, '/')
     await expect.poll(() => menuOptions(page), { timeout: 15_000 })
       .toEqual(expect.arrayContaining([expect.stringContaining(SKILL_NAME)]))
     const onStandard = await menuOptions(page)
     expect(onStandard.some(option => option.startsWith('compact'))).toBe(true)
     expect(onStandard.some(option => option.startsWith('goal'))).toBe(true)
     expect(onStandard.some(option => option.startsWith('plan'))).toBe(true)
-    await composer.fill('')
+    await writeComposerDraft(page, composer, '')
   }, 90_000)
 
   it('labels a resumed session with the preset it was created under', async () => {

apps/web/tests/expected/agent-preset-selection/menu.expected.md

@@ -5,3 +5,4 @@
   - menuitem "PTC mode All Standard mode capabilities, with tools exposed through the Code Mode SDK so the model can combine multi-step operations in one TypeScript program."
   - menuitem "Minimal mode Two-tool coding agent with persistent bash and str_replace_editor."
   - menuitem "Creator mode Built for creating custom agent presets, with all Standard mode capabilities plus runtime inspection, plugin experiments, and preset-authoring guidance."
+  - menuitem "Refusing mode Resolves, then refuses to start."

apps/web/tests/subagent-conversation.e2e.ts

@@ -267,7 +267,7 @@ describe('web e2e: persisted subagent conversation and human continuation', () =
 
   it('keeps known descendants reachable across a stale empty catalog response', async () => {
     onTestFailed(() => saveFailureShot(page, 'web-e2e-subagent-stale-catalog'))
-    const pattern = '**/api/subagent.list'
+    const pattern = '**/api/subagents/list'
     let firstClaimed = false
     let emptyDelivered = false
     let trailingRequested = false
@@ -373,7 +373,7 @@ describe('web e2e: persisted subagent conversation and human continuation', () =
 
   it('keeps a restored child neutral until its parent availability arrives', async () => {
     onTestFailed(() => saveFailureShot(page, 'web-e2e-subagent-restore'))
-    const pattern = '**/api/subagent.list'
+    const pattern = '**/api/subagents/list'
     let requested = false
     let releaseCatalog = (): void => {}
     const catalogHeld = new Promise<void>((resolve) => { releaseCatalog = resolve })
@@ -546,7 +546,7 @@ describe('web e2e: persisted subagent conversation and human continuation', () =
     const input = page.locator('[data-composer-input][contenteditable="true"]').first()
     await input.waitFor()
     const promptResponse = page.waitForResponse(response =>
-      new URL(response.url()).pathname === '/api/subagent.prompt')
+      new URL(response.url()).pathname === '/api/subagents/prompt')
     await input.fill(POST_FORK_FOLLOWUP)
     await input.press('Enter')
     const promptReceipt = await (await promptResponse).json() as {

apps/web/tests/subagent-interrupt-ui.e2e.ts

@@ -1,7 +1,7 @@
 // Web e2e scenario: the composer's independent Stop interrupts a running
 // continuable child. The child holds its model turn open through a replay
 // hang entry; the browser proves Send and Stop coexist, the parent-offline
-// disabled-Send-with-Stop composer, the subagent.interrupt
+// disabled-Send-with-Stop composer, the subagents/interruptByParent
 // (never session.cancel) transport, the parked follow-up, and the FIFO resume
 // on a waking send.
 //
@@ -187,7 +187,7 @@ describe.skipIf(MODE === 'record')('web e2e: composer interrupt for a running co
     // parentAvailable: false while the child Activation stays live (the
     // interrupt RPC itself needs no live parent — covered host-side by
     // subagent-interrupt.e2e.ts).
-    const pattern = '**/api/subagent.list'
+    const pattern = '**/api/subagents/list'
     await page.route(pattern, async (route) => {
       const response = await route.fetch()
       const body = await response.json() as {
@@ -228,7 +228,7 @@ describe.skipIf(MODE === 'record')('web e2e: composer interrupt for a running co
       )
       const aborted = waitForAbortedTurn(scaffold, childId)
       const interruptResponse = page.waitForResponse(response =>
-        new URL(response.url()).pathname === '/api/subagent.interrupt')
+        new URL(response.url()).pathname === '/api/subagents/interruptByParent')
       await stop.click()
       expect(((await (await interruptResponse).json()) as {
         result: { ok: boolean; value?: { accepted: boolean } }
@@ -252,7 +252,7 @@ describe.skipIf(MODE === 'record')('web e2e: composer interrupt for a running co
     }
   }, 60_000)
 
-  it('interrupts through subagent.interrupt, parks the follow-up, and resumes it FIFO', async () => {
+  it('interrupts through subagents/interruptByParent, parks the follow-up, and resumes it FIFO', async () => {
     onTestFailed(() => saveFailureShot(page, 'web-e2e-subagent-interrupt-flow'))
     // Reselect the child with the truthful catalog: parent available again.
     await page.getByRole('navigation', { name: 'Session hierarchy' })
@@ -265,7 +265,7 @@ describe.skipIf(MODE === 'record')('web e2e: composer interrupt for a running co
 
     // Queue a follow-up through Send while independent Stop remains available.
     const promptResponse = page.waitForResponse(response =>
-      new URL(response.url()).pathname === '/api/subagent.prompt')
+      new URL(response.url()).pathname === '/api/subagents/prompt')
     await input.fill(FOLLOWUP)
     await page.getByRole('button', { name: 'Send message' }).click()
     expect(((await (await promptResponse).json()) as { result: { ok: boolean } }).result)
@@ -275,7 +275,7 @@ describe.skipIf(MODE === 'record')('web e2e: composer interrupt for a running co
     const stop = page.getByRole('button', { name: 'Stop generating' })
     expect(await stop.count()).toBe(1)
     const interruptResponse = page.waitForResponse(response =>
-      new URL(response.url()).pathname === '/api/subagent.interrupt')
+      new URL(response.url()).pathname === '/api/subagents/interruptByParent')
     await stop.click()
     expect(((await (await interruptResponse).json()) as {
       result: { ok: boolean; value?: { accepted: boolean } }

apps/web/tests/subagent-interrupt.e2e.ts

@@ -1,4 +1,4 @@
-// Web e2e scenario (browserless): the subagent.interrupt RPC against the real
+// Web e2e scenario (browserless): the subagents interrupt Remote against the real
 // composition. A live continuable child holds its model turn open through a
 // replay hang entry; plain HTTP queues a follow-up, interrupts the turn, and
 // proves from the real session state that the turn aborted, the follow-up
@@ -22,25 +22,12 @@ const WAKING = 'And add one concrete example.'
 
 type RpcResult<T> = { ok: true; value: T } | { ok: false; error: { code: string; message: string } }
 
-/** POST one API Proxy unary RPC through the real HTTP carrier and unwrap its result. */
-async function rpc<T>(scaffold: WebScaffold, method: string, payload: unknown): Promise<RpcResult<T>> {
-  const response = await scaffold.hostFetch(`/api/${method}`, {
-    method: 'POST',
-    headers: { 'content-type': 'application/json' },
-    body: JSON.stringify({
-      type: 'client-request',
-      rpcId: `interrupt-e2e-${method}-${randomUUID()}`,
-      method,
-      payload,
-    }),
-  })
-  if (!response.ok) throw new Error(`${method} failed over HTTP ${response.status}: ${await response.text()}`)
-  return (await response.json() as { result: RpcResult<T> }).result
-}
-
-/** POST one generated Session Remote unary through the API Gateway carrier. */
-async function sessionRemote<T>(scaffold: WebScaffold, method: string, request: unknown): Promise<RpcResult<T>> {
-  const endpoint = `session/${method}`
+/** POST one generated Remote unary through the API Gateway carrier. */
+async function remote<T>(
+  scaffold: WebScaffold,
+  endpoint: string,
+  args: Readonly<Record<string, unknown>>,
+): Promise<RpcResult<T>> {
   const response = await scaffold.hostFetch(`/api/${endpoint}`, {
     method: 'POST',
     headers: { 'content-type': 'application/json' },
@@ -48,13 +35,18 @@ async function sessionRemote<T>(scaffold: WebScaffold, method: string, request:
       type: 'client-request',
       rpcId: `interrupt-e2e-${endpoint}-${randomUUID()}`,
       method: endpoint,
-      payload: { args: { request } },
+      payload: { args },
     }),
   })
   if (!response.ok) throw new Error(`${endpoint} failed over HTTP ${response.status}: ${await response.text()}`)
   return (await response.json() as { result: RpcResult<T> }).result
 }
 
+/** POST one generated Session Remote unary through the API Gateway carrier. */
+function sessionRemote<T>(scaffold: WebScaffold, method: string, request: unknown): Promise<RpcResult<T>> {
+  return remote<T>(scaffold, `session/${method}`, { request })
+}
+
 /** Poll a synchronous condition (hook-safe; expect.poll is test-body only). */
 async function waitFor(predicate: () => boolean, what: string, timeoutMs = 30_000): Promise<void> {
   const deadline = Date.now() + timeoutMs
@@ -78,7 +70,7 @@ function textCompletion(text: string): object {
   }
 }
 
-describe.skipIf(MODE === 'record')('web e2e: subagent.interrupt over the real composition', () => {
+describe.skipIf(MODE === 'record')('web e2e: subagents/interruptByParent over the real composition', () => {
   let scaffold: WebScaffold
   let sidecarRoot: string
   let readyFile: string
@@ -138,18 +130,21 @@ describe.skipIf(MODE === 'record')('web e2e: subagent.interrupt over the real co
 
   it('parks a queued follow-up on interrupt and resumes it FIFO on a waking send', async () => {
     // Queue the follow-up while the turn is still open, then interrupt.
-    const queued = await rpc<{ messageId: string }>(scaffold, 'subagent.prompt', {
-      parentSessionId: parentId,
-      childSessionId: childId,
-      mode: 'continuable',
-      content: [{ type: 'text', text: FOLLOWUP }],
+    const queued = await remote<{ messageId: string }>(scaffold, 'subagents/prompt', {
+      request: {
+        requestId: randomUUID(),
+        parentSessionId: parentId,
+        childSessionId: childId,
+        mode: 'continuable',
+        content: [{ type: 'text', text: FOLLOWUP }],
+      },
     })
     expect(queued).toMatchObject({ ok: true })
 
     const settled = scaffold.whenTurnSettled()
-    const interrupted = await rpc<{ accepted: true }>(scaffold, 'subagent.interrupt', {
-      parentSessionId: parentId,
+    const interrupted = await remote<{ accepted: true }>(scaffold, 'subagents/interruptByParent', {
       childSessionId: childId,
+      parentSessionId: parentId,
       mode: 'continuable',
     })
     expect(interrupted).toMatchObject({ ok: true, value: { accepted: true } })
@@ -169,11 +164,14 @@ describe.skipIf(MODE === 'record')('web e2e: subagent.interrupt over the real co
 
     // Only an explicit waking send resumes the parked queue, FIFO, then the
     // child runs both turns to completion and settles.
-    const waking = await rpc<{ messageId: string }>(scaffold, 'subagent.prompt', {
-      parentSessionId: parentId,
-      childSessionId: childId,
-      mode: 'continuable',
-      content: [{ type: 'text', text: WAKING }],
+    const waking = await remote<{ messageId: string }>(scaffold, 'subagents/prompt', {
+      request: {
+        requestId: randomUUID(),
+        parentSessionId: parentId,
+        childSessionId: childId,
+        mode: 'continuable',
+        content: [{ type: 'text', text: WAKING }],
+      },
     })
     expect(waking).toMatchObject({ ok: true })
     await expect.poll(() => scaffold.ctx.agents.get(childId), { timeout: 60_000 }).toBeUndefined()

apps/web/tests/trajectory-image-display.expected.e2e.ts

@@ -6,12 +6,24 @@
 // selecting the ledger record renders the shared ui-attachment gallery from
 // the durable session-log reference, and the browser URL is the SAME object
 // URL Chat resolved — one sessions.attachment read per session attachment.
-import { fireEvent, screen, waitFor, within } from '@testing-library/react'
+import { act, fireEvent, screen, waitFor, within } from '@testing-library/react'
 import { expect, it, vi } from 'vitest'
 import { installAssembledBootEnv, mountAssembledApp } from './assembled-boot.ts'
 
 installAssembledBootEnv()
 
+/**
+ * How long the mounted tree waits out the virtual ledger's scroll-idle timer.
+ * jsdom fires no `scrollend`, so `@tanstack/react-virtual` falls back to a
+ * debounce it re-arms on every scroll event (`isScrollingResetDelay`, 150ms by
+ * default) and its unsubscribe removes only the listeners; a scenario that
+ * ends inside that window leaves the timer to re-render the table after vitest
+ * has torn this file's jsdom down, where React reads a `window` that is gone.
+ * Armed later and with a longer delay than the debounce, this wait always
+ * expires after it.
+ */
+const SCROLL_IDLE_DRAIN_MS = 400
+
 /** Open the fixture history session and wait for the Chat gallery to load. */
 async function openFixtureSession(): Promise<void> {
   const tree = await screen.findByRole('tree', { name: 'Sessions' }, { timeout: 10_000 })
@@ -41,16 +53,24 @@ async function scrollRowIntoWindow(needle: string): Promise<HTMLElement> {
   }, { timeout: 10_000 })
   const pane = document.querySelector('[data-trajectory-scroll] table')?.parentElement
   if (!(pane instanceof HTMLElement)) throw new Error('trajectory scroll pane missing')
-  for (let top = 0; top <= 40_000; top += 1_000) {
+  const findRow = (): HTMLElement | undefined =>
+    [...document.querySelectorAll<HTMLElement>('tr[data-trajectory-row-key]')]
+      .find(row => row.textContent?.includes(needle))
+  let mounted = false
+  for (let top = 0; !mounted && top <= 40_000; top += 1_000) {
     pane.scrollTop = top
     fireEvent.scroll(pane)
     // Let the virtualizer publish the new window before probing.
     await new Promise(resolve => setTimeout(resolve, 25))
-    const hit = [...document.querySelectorAll<HTMLElement>('tr[data-trajectory-row-key]')]
-      .find(row => row.textContent?.includes(needle))
-    if (hit !== undefined) return hit
+    mounted = findRow() !== undefined
   }
-  throw new Error(`trajectory row containing ${JSON.stringify(needle)} never mounted`)
+  // Nothing scrolls the ledger after this, so draining the scroll-idle
+  // debounce here leaves no timer armed for the rest of the scenario. The
+  // drained reset re-renders the window, so the row is read afterwards.
+  await act(async () => { await new Promise(resolve => setTimeout(resolve, SCROLL_IDLE_DRAIN_MS)) })
+  const hit = findRow()
+  if (hit === undefined) throw new Error(`trajectory row containing ${JSON.stringify(needle)} never mounted`)
+  return hit
 }
 
 it('renders durable record images in the Trajectory details panel from the shared cache', async () => {