test: stabilize Cordis replay and Windows paths

apps/web/tests/cordis-tool-round.e2e.ts

@@ -108,8 +108,10 @@ describe('web e2e: Cordis tools use their owned cards', () => {
     // NOT the plugin running. Until a person answers, the browser half has not
     // been fetched, evaluated, or mounted anywhere on this page.
     expect(await page.locator('[data-snapshot-probe]').count()).toBe(0)
+    const approvalTurnSettled = scaffold.whenTurnSettled()
     await approve.click()
     await expect.poll(() => page.locator('[data-snapshot-probe]').count(), { timeout: 30_000 }).toBe(1)
+    await approvalTurnSettled
 
     const sessionId = await runTurnSettled
     const stopTurnSettled = scaffold.whenTurnSettled()
@@ -162,6 +164,12 @@ describe('web e2e: Cordis tools use their owned cards', () => {
 
   it.skipIf(MODE === 'record')('matches the conversation aria golden', async () => {
     onTestFailed(() => saveFailureShot(page, 'web-e2e-cordis-aria'))
+    await page.locator('[data-conversation-scroll]').evaluate((host) => { host.scrollTop = host.scrollHeight })
+    await expect.poll(
+      async () => page.getByRole('button', { name: 'Back to bottom', exact: true }).count(),
+      { timeout: 10_000 },
+    ).toBe(0)
+    await page.mouse.move(0, 0)
     const snapshot = await captureStableAria(page, '[class*="centerCol"]', scaffold.workspaceCwd)
     await compareOrRefreshGolden(UI_EXPECTED, snapshot, MODE)
   })

apps/web/tests/snapshots/cordis-tool-round/session.jsonl

@@ -47,11 +47,15 @@
 {"type":"step/end","seq":248,"time":1785157569554,"data":{"turn":1,"step":4}}
 {"type":"turn/end","seq":249,"time":1785157569554,"data":{"turn":1,"reason":{"kind":"completed"}}}
 {"type":"user/message","seq":250,"time":1785157569555,"data":{"content":[{"type":"text","text":"Use only Cordis tools. Call cordis_stop with pluginId \"snap-1\". After it succeeds, reply exactly CORDIS_UI_DONE and stop."}],"source":{"kind":"user","rpcId":"{{rpcId}}"}},"surfaceOp":"append"}
-{"type":"assistant/chunk","seq":251,"time":1785157569556,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
-{"type":"assistant/chunk","seq":252,"time":1785157569557,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"cordis-stop","name":"cordis_stop","arguments":"{\"pluginId\":\"snap-1\"}"}}}}
+{"type":"assistant/chunk","seq":251,"time":1785157569556,"data":{"turn":2,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":252,"time":1785157569557,"data":{"turn":2,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"The Cordis Plugin is running."}}}}
 {"type":"assistant/chunk","seq":253,"time":1785157569558,"data":{"turn":2,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
-{"type":"assistant/chunk","seq":254,"time":1785157569559,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
-{"type":"assistant/chunk","seq":255,"time":1785157569560,"data":{"turn":2,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
-{"type":"assistant/chunk","seq":256,"time":1785157569561,"data":{"turn":2,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CORDIS_UI_DONE"}}}}
-{"type":"assistant/chunk","seq":257,"time":1785157569562,"data":{"turn":2,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
-{"type":"assistant/chunk","seq":258,"time":1785157569563,"data":{"turn":2,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/chunk","seq":254,"time":1785157569559,"data":{"turn":2,"step":1,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}
+{"type":"assistant/chunk","seq":255,"time":1785157569560,"data":{"turn":3,"step":1,"chunk":{"type":"block-start","index":0,"blockType":"tool-call"}}}
+{"type":"assistant/chunk","seq":256,"time":1785157569561,"data":{"turn":3,"step":1,"chunk":{"type":"block-end","index":0,"block":{"type":"tool-call","id":"cordis-stop","name":"cordis_stop","arguments":"{\"pluginId\":\"snap-1\"}"}}}}
+{"type":"assistant/chunk","seq":257,"time":1785157569562,"data":{"turn":3,"step":1,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
+{"type":"assistant/chunk","seq":258,"time":1785157569563,"data":{"turn":3,"step":1,"chunk":{"type":"finish","reason":{"kind":"tool-calls"}}}}
+{"type":"assistant/chunk","seq":259,"time":1785157569564,"data":{"turn":3,"step":2,"chunk":{"type":"block-start","index":0,"blockType":"text"}}}
+{"type":"assistant/chunk","seq":260,"time":1785157569565,"data":{"turn":3,"step":2,"chunk":{"type":"block-end","index":0,"block":{"type":"text","text":"CORDIS_UI_DONE"}}}}
+{"type":"assistant/chunk","seq":261,"time":1785157569566,"data":{"turn":3,"step":2,"chunk":{"type":"usage","usage":{"inputTokens":3,"outputTokens":3}}}}
+{"type":"assistant/chunk","seq":262,"time":1785157569567,"data":{"turn":3,"step":2,"chunk":{"type":"finish","reason":{"kind":"stop"}}}}

apps/web/tests/snapshots/cordis-tool-round/ui.expected.md

@@ -66,6 +66,18 @@
   - img
   - img
   - text: Context injection cordis-host-runner
+- paragraph: The Cordis Plugin is running.
+- button "Copy":
+  - img
+- button "Good response":
+  - img
+- button "Bad response":
+  - img
+- button "Branch into a new conversation":
+  - img
+- text: {{clock}} Ran for {{duration}} Use only Cordis tools. Call cordis_stop with pluginId "snap-1". After it succeeds, reply exactly CORDIS_UI_DONE and stop. {{clock}}
+- button "Copy":
+  - img
 - img
 - text: Stop Cordis Plugin snap-1
 - button "Inspect"
@@ -79,14 +91,7 @@
   - img
 - button "Branch into a new conversation":
   - img
-- text: {{clock}} Ran for {{duration}} Use only Cordis tools. Call cordis_stop with pluginId "snap-1". After it succeeds, reply exactly CORDIS_UI_DONE and stop. {{clock}}
-- button "Copy":
-  - img
-- status:
-  - text: "This turn failedllm-replay: script exhausted — session requested model call #7 but its script has only 6; re-record the scenario"
-  - code: UNKNOWN
-- button "Back to bottom":
-  - img
+- text: {{clock}} Ran for {{duration}}
 - textbox "Message the agent"
 - button "Commands":
   - img
@@ -96,4 +101,4 @@
   - img
 - button "0% of context used"
 - button "Send message" [disabled]
-- text: 3 turns · 7 steps LLM {{duration}} · Tool call {{duration}} Cache hit 77% Input 66.5K tok · Output 318 tok
+- text: 3 turns · 7 steps LLM {{duration}} · Tool call {{duration}} Cache hit 77% Input 66.5K tok · Output 321 tok