fix(story-mode): name the check that failed, say whose plan a failure was, and when background work finished

From review round 9:
- "another check" was too vague to act on. A check other than the tests
  is named by its tool: a layout check (a formatter), a style check (a
  linter), a type check, the build. "Fix and test, 2 rounds: a test fails
  and Claude fixes it, then a layout check fails and Claude fixes that;
  everything passes."
- A failure the prompt asked for reads "The first run fails on purpose,
  as you asked; the second passes." ("The tests don't even load, as you
  asked."); one Claude planned itself keeps "Test first: … as planned".
- A test run sent to the background that reported in a later turn reads
  "Claude starts the tests in the background; later they pass.", not as
  if they passed in that turn.
- The test board says "1 test" and, when nothing loaded, "1 error".

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

src/story-mode.ts

@@ -2259,14 +2259,26 @@ function printableLine(line: string): string | null {
   return clip(text.replace(/[\w.+-]+@[\w-]+(\.[\w-]+)+/g, '…@…').replace(/[A-Za-z0-9+/=_-]{32,}/g, '…'), 90);
 }
 
+/** What a check other than the tests is, by its tool, for a reader: "a layout check", "a style check". */
+function checkName(e: StoryEvent | undefined): string {
+  const cmd = e?.label ?? '';
+  if (/\b(ruff format|black|prettier|gofmt|rustfmt|isort|autopep8|biome format)\b/.test(cmd)) return 'a layout check';
+  if (/\b(ruff|eslint|flake8|pylint|clippy|shellcheck|golangci-lint|stylelint|biome)\b/.test(cmd))
+    return 'a style check';
+  if (/\b(tsc|mypy|pyright|vue-tsc)\b/.test(cmd)) return 'a type check';
+  if (/\b(build|make|cargo build|go build|webpack|vite build|esbuild)\b/.test(cmd)) return 'the build';
+  return 'another check';
+}
+
 /** You asked for a failing run first (test first), or Claude said the failure was expected. */
 const PLANNED_IN_PROMPT =
   /\b(test[- ]first|tdd|fail(s|ing)? (first|on purpose)|failing tests? first|on purpose|red first|before (the )?(code|module|function|implementation) exists)\b/i;
 const EXPECTED_BY_CLAUDE =
   /\b(as expected|as planned|as intended|expected (failure|to fail|red)|deliberately fail\w*|on purpose|test[- ]first|the red state)\b/i;
-function plannedFailure(turn: StoryTurn, events: StoryEvent[]): boolean {
+function plannedFailure(turn: StoryTurn, events: StoryEvent[]): false | 'asked' | 'planned' {
   const prompt = events.find((e) => e.intent === 'prompt')?.label ?? turn.prompt;
-  return PLANNED_IN_PROMPT.test(prompt) || events.some((e) => e.intent === 'say' && EXPECTED_BY_CLAUDE.test(e.label));
+  if (PLANNED_IN_PROMPT.test(prompt)) return 'asked';
+  return events.some((e) => e.intent === 'say' && EXPECTED_BY_CLAUDE.test(e.label)) ? 'planned' : false;
 }
 
 function decorate(draft: DraftPanel, events: StoryEvent[], turn: StoryTurn, index: number): StoryPanel {
@@ -2451,27 +2463,35 @@ function decorate(draft: DraftPanel, events: StoryEvent[], turn: StoryTurn, inde
         const n = panel.rounds;
         const testFails = fails - lintFails;
         // A failure you asked for (test first), or that Claude said it expected, is the plan working.
-        const planned = plannedFailure(turn, events) && passed(checks[0]) === false;
+        const planned = passed(checks[0]) === false ? plannedFailure(turn, events) : false;
+        // The check that failed while the tests passed, by name: "a layout check", not "another check".
+        const otherCheck = checkName(
+          failed.find((e) => !!e.tests && e.tests.passed > 0 && !e.tests.failed && !e.tests.errors)
+        );
         story =
           panel.ok === true && !fails
             ? `Build and test, ${plural(n, 'round')}. Every check passes.`
             : panel.ok === true && fails === lintFails
-              ? `Build and test, ${plural(n, 'round')}. The tests pass every time; another check fails ${fails === 1 ? 'once' : `${fails} times`}, then passes.`
-              : panel.ok === true && planned && checks.length === 2
-                ? 'Test first: the first run fails as planned, the second passes.'
-                : panel.ok === true && planned
-                  ? `Test first, ${plural(n, 'round')}: the first run fails as planned, then it passes.`
-                  : panel.ok === true && checks.length === 2 && fails === 1
-                    ? 'Fix and test: the first run fails, the second passes.'
-                    : panel.ok === true && testFails && lintFails
-                      ? `Fix and test, ${plural(n, 'round')}: ${testFails === 1 ? 'a test run fails' : `${testFails} test runs fail`}, then another check, then everything passes.`
-                      : panel.ok === true
-                        ? `Fix and test, ${plural(n, 'round')}: ${fails === 1 ? 'one run fails' : `${fails} runs fail`}, then it passes.`
-                        : panel.ok === false && tests && tests.passed && !tests.failed && !tests.errors
-                          ? `Fix and test, ${plural(n, 'round')}. The tests pass, but another check fails.`
-                          : panel.ok === false
-                            ? `Fix and test, ${plural(n, 'round')}. Still failing.`
-                            : `Fix and test, ${plural(n, 'round')}.`;
+              ? `Build and test, ${plural(n, 'round')}. The tests pass every time; ${otherCheck} fails ${fails === 1 ? 'once' : `${fails} times`}, then passes.`
+              : panel.ok === true && planned === 'asked'
+                ? checks.length === 2
+                  ? 'The first run fails on purpose, as you asked; the second passes.'
+                  : `${plural(n, 'round')}: the first run fails on purpose, as you asked, then it passes.`
+                : panel.ok === true && planned && checks.length === 2
+                  ? 'Test first: the first run fails as planned, the second passes.'
+                  : panel.ok === true && planned
+                    ? `Test first, ${plural(n, 'round')}: the first run fails as planned, then it passes.`
+                    : panel.ok === true && checks.length === 2 && fails === 1
+                      ? 'Fix and test: the first run fails, the second passes.'
+                      : panel.ok === true && testFails && lintFails
+                        ? `Fix and test, ${plural(n, 'round')}: ${testFails === 1 ? 'a test fails' : `tests fail ${testFails} times`} and Claude fixes it, then ${otherCheck} fails and Claude fixes that; everything passes.`
+                        : panel.ok === true
+                          ? `Fix and test, ${plural(n, 'round')}: ${fails === 1 ? 'one run fails' : `${fails} runs fail`}, then it passes.`
+                          : panel.ok === false && tests && tests.passed && !tests.failed && !tests.errors
+                            ? `Fix and test, ${plural(n, 'round')}. The tests pass, but another check fails.`
+                            : panel.ok === false
+                              ? `Fix and test, ${plural(n, 'round')}. Still failing.`
+                              : `Fix and test, ${plural(n, 'round')}.`;
         // Commits made inside the loop (a test one-liner that ends in one) are part of its story.
         const saved = evs.reduce((k, e) => k + (e.commits ?? 0), 0);
         if (saved) story += ` Claude saves the work${saved > 1 ? ` in ${saved} commits` : ''} along the way.`;
@@ -2491,9 +2511,11 @@ function decorate(draft: DraftPanel, events: StoryEvent[], turn: StoryTurn, inde
       const failedFirst = checks.slice(0, -1).some((e) => passed(e) === false);
       if (lastCheck?.running)
         story = isTest ? 'Claude starts the tests in the background.' : 'Claude starts the checks in the background.';
-      // Finished later, in the background: not tests that ran during this turn.
+      // Finished in the background: not tests that ran during this turn, and said when it was later.
       else if (lastCheck?.background && ok !== undefined)
-        story = `The ${isTest ? 'tests' : 'checks'} run in the background and ${ok ? 'pass' : 'fail'}.`;
+        story = lastCheck.finishedLater
+          ? `Claude starts the ${isTest ? 'tests' : 'checks'} in the background; later they ${ok ? 'pass' : 'fail'}.`
+          : `The ${isTest ? 'tests' : 'checks'} run in the background and ${ok ? 'pass' : 'fail'}.`;
       else if (ok === undefined) story = isTest ? 'Claude runs the tests.' : 'Claude runs the checks.';
       else if (!ok && tests && !tests.passed && !tests.failed && tests.errors) story = "The tests don't even load.";
       else if (!ok && tests && tests.passed && !tests.failed && !tests.errors)
@@ -2506,7 +2528,9 @@ function decorate(draft: DraftPanel, events: StoryEvent[], turn: StoryTurn, inde
             ? `${tests.failed} tests fail.`
             : 'A test fails.'
           : 'A check fails.';
-      if (ok === false && plannedFailure(turn, events)) story = `${story.replace(/[.]$/, '')}, as planned.`;
+      const plannedCheck = ok === false ? plannedFailure(turn, events) : false;
+      if (plannedCheck)
+        story = `${story.replace(/[.]$/, '')}, ${plannedCheck === 'asked' ? 'as you asked' : 'as planned'}.`;
       // The same command with the same outcome twice in a row reads once, with a count.
       const runs: Array<{ text: string; n: number }> = [];
       for (const e of checks) {

src/web/public/story-mode.js

@@ -689,6 +689,11 @@ const StoryArt = (() => {
     return { w, h };
   }
   /** A terminal card: the commands (`› …`), then, dimmer, the first lines one of them printed. */
+  /** A test board's label: "1 test", "19 tests", or "1 error" when nothing even loaded. */
+  function boardLabel(t, total) {
+    if (t && !t.passed && !t.failed && t.errors) return `${t.errors} error${t.errors === 1 ? '' : 's'}`;
+    return `${total} test${total === 1 ? '' : 's'}`;
+  }
   function terminal(ctx, x, y, w, h, lines, f, printed = []) {
     rr(ctx, x, y, w, h, 7);
     inked(ctx, C.line, 2.5);
@@ -1094,7 +1099,7 @@ const StoryArt = (() => {
           const dots = Math.min(42, total);
           const fail = bad ? Math.max(1, Math.round((bad / total) * dots)) : 0;
           const fill = running ? (f * 0.6) % (dots + 6) : Math.min(dots, (f - 4) * 1.6);
-          b = board(ctx, 16, 106, `${total} tests`, dots, fail, fill);
+          b = board(ctx, 16, 106, boardLabel(t, total), dots, fail, fill);
         } else {
           b = verdict(ctx, 16, 106, 'checks', p.panel.ok, f);
         }
@@ -1153,7 +1158,7 @@ const StoryArt = (() => {
           if (total) {
             const dots = Math.min(28, total);
             const fail = bad ? Math.max(1, Math.round((bad / total) * dots)) : 0;
-            b = board(ctx, bx, 116, `${total} tests`, dots, fail, Math.min(dots, (f - 4) * 1.6));
+            b = board(ctx, bx, 116, boardLabel(t, total), dots, fail, Math.min(dots, (f - 4) * 1.6));
           } else b = verdict(ctx, bx, 116, 'checks', p.panel.ok, f);
           hit('output', bx, 116, b.w, b.h, 'See the test output');
         }

test/fixtures/story-corpus/background-helpers-reports.expected.json

@@ -221,7 +221,7 @@
       {
         "kind": "check",
         "ok": true,
-        "story": "The tests run in the background and pass.",
+        "story": "Claude starts the tests in the background; later they pass.",
         "director": "python3 -m pytest -q → ok",
         "balloon": null,
         "tests": null,

test/fixtures/story-corpus/errors-push-append.expected.json

@@ -73,7 +73,7 @@
       {
         "kind": "check",
         "ok": false,
-        "story": "The tests don't even load, as planned.",
+        "story": "The tests don't even load, as you asked.",
         "director": "python3 -m pytest -q → error · 0 passed, 1 error",
         "balloon": "Now the red run, once, with calc_stats.py confirmed absent:",
         "tests": {

test/fixtures/story-corpus/failing-test-on-purpose.expected.json

@@ -51,7 +51,7 @@
       {
         "kind": "loop",
         "ok": true,
-        "story": "Test first: the first run fails as planned, the second passes.",
+        "story": "The first run fails on purpose, as you asked; the second passes.",
         "director": "2 checks · 3 edits · +35 −1 · last: ok",
         "balloon": "All 4 pass.",
         "tests": {
@@ -133,7 +133,7 @@
       {
         "kind": "loop",
         "ok": true,
-        "story": "Test first: the first run fails as planned, the second passes.",
+        "story": "The first run fails on purpose, as you asked; the second passes.",
         "director": "2 checks · 2 edits · +5 −1 · last: ok",
         "balloon": "All 5 pass.",
         "tests": {

test/fixtures/story-corpus/long-turn-prompt-added.expected.json

@@ -73,7 +73,7 @@
       {
         "kind": "loop",
         "ok": true,
-        "story": "Build and test, 4 rounds. The tests pass every time; another check fails once, then passes.",
+        "story": "Build and test, 4 rounds. The tests pass every time; a layout check fails once, then passes.",
         "director": "5 checks · 10 edits · +694 · last: ok",
         "balloon": "All 4 modules are done: 63 tests pass and ruff is clean.",
         "tests": {

test/fixtures/story-corpus/python-cli-fix-loop.expected.json

@@ -51,7 +51,7 @@
       {
         "kind": "loop",
         "ok": true,
-        "story": "Fix and test, 2 rounds: a test run fails, then another check, then everything passes. Claude saves the work in 4 commits along the way.",
+        "story": "Fix and test, 2 rounds: a test fails and Claude fixes it, then a layout check fails and Claude fixes that; everything passes. Claude saves the work in 4 commits along the way.",
         "director": "3 checks · 5 edits · +218 −1 · last: ok",
         "balloon": "All 19 tests pass, ruff check is clean, and the output matches real wc column for column.",
         "tests": {

test/story-corpus.test.ts

@@ -77,7 +77,7 @@ describe('Story Mode on real sessions', () => {
     expect(panel('python-cli-fix-loop', 0, 'loop')).toMatchObject({
       // The failing test was a real bug, the second failure only ruff: said apart, with the 4 commits made inside.
       story:
-        'Fix and test, 2 rounds: a test run fails, then another check, then everything passes. Claude saves the work in 4 commits along the way.',
+        'Fix and test, 2 rounds: a test fails and Claude fixes it, then a layout check fails and Claude fixes that; everything passes. Claude saves the work in 4 commits along the way.',
       roundsOk: [false, false, true],
     });
   });
@@ -131,7 +131,7 @@ describe('Story Mode on real sessions', () => {
 
   it('tests that do not load, a commit then a push in its own command, a push with a failing step after it', () => {
     const [first, second, third] = pagesOf('errors-push-append');
-    expect(first.panels.find((p) => p.kind === 'check')?.story).toBe("The tests don't even load, as planned.");
+    expect(first.panels.find((p) => p.kind === 'check')?.story).toBe("The tests don't even load, as you asked.");
     expect(second.panels.find((p) => p.kind === 'loop')).toMatchObject({
       story: 'Test first: the first run fails as planned, the second passes.',
       roundsOk: [false, true],
@@ -155,7 +155,9 @@ describe('Story Mode on real sessions', () => {
       expect(report.helpers).toBe(0);
       expect(report.panels[0].story).toBe('A background job reports back.');
     }
-    expect(pages[3].panels.map((p) => p.story)).toContain('The tests run in the background and pass.');
+    expect(pages[3].panels.map((p) => p.story)).toContain(
+      'Claude starts the tests in the background; later they pass.'
+    );
     // A python script that rewrote README.md counts as a change, beside the two files written.
     expect(pages[3].panels.map((p) => p.story)).toContain('Claude changes 3 files.');
   });

test/story-mode.test.ts

@@ -742,7 +742,7 @@ describe('buildStoryTurns', () => {
     );
     // Not tests that ran during that turn: they ran later, in the background.
     expect(first.panels.find((p) => p.kind === 'check')!.caption.story).toBe(
-      'The tests run in the background and pass.'
+      'Claude starts the tests in the background; later they pass.'
     );
     const second = buildStoryPage(turns[1]);
     expect(second.panels.find((p) => p.kind === 'check')).toMatchObject({
@@ -1194,7 +1194,7 @@ describe('buildStoryPage', () => {
     const loop = page.panels.find((p) => p.kind === 'loop')!;
     expect(loop.roundsOk).toEqual([true, false, true]);
     expect(loop.caption.story).toBe(
-      'Build and test, 2 rounds. The tests pass every time; another check fails once, then passes.'
+      'Build and test, 2 rounds. The tests pass every time; a layout check fails once, then passes.'
     );
   });