feat(story-mode): a run panel shows what its command printed

From review round 8: you asked to see a dry run, Claude's balloon said
"Here's the dry run you asked for", and the page showed nothing of it.
A run panel's terminal card now prints, dimmer under the commands, the
first lines its last command printed (the last ones when it failed:
that is where an error is).

The card is drawn on the canvas, so it goes into a saved image someone
may share: a line that looks like a secret (token, password, api key,
credential, inside identifiers too) is left out, addresses and long
token-like strings are masked, and Claude Code's own markers ("(Bash
completed with no output)", tool-error tags) are dropped.

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

docs/api-reference.md

@@ -1083,7 +1083,7 @@ A Bash call sent to the background (its result is only `Command running in backg
 A `page` carries the turn's summary fields plus:
 
 - `events`: one per prompt, sentence Claude wrote and tool call. Each has `intent`, `tool`, `label` and, when the call returned, `ok` (`false` when the result was flagged as an error, or reported failed tests or lint problems a pipe kept from the exit code). Depending on the call: `edit` (an Edit/Write, hunks in Claude Code's own `structuredPatch` shape), `edits` (files a Bash call wrote through here-documents or `printf`/`echo` redirects, that a python or node script fed by a here-document opens for writing, or that Claude Code says it changed; the last two have no lines, since what changed is not in the command; an append's hunk has `atEnd: true` and no line numbers, since where it lands is not in the command), the tail of `output`, `tests` (`{ passed, failed, errors? }`, counted on the whole output before it is cut), `helper`, `files` (files a Bash call printed with `cat`/`head`/`tail`, resolved against the command's folder), `headline` (the part of a one-liner that decided its intent), `commits`, `subject` (the first commit's message, when the command spells it out with `-m` or a `git commit -F -` here-document), `ships` (`push` / `pr`), `running`, `background` and `finishedLater` (see Background work), `keyPoints` (for a sentence Claude wrote: up to two sentences from further into a long message that a reader must not lose, such as an open problem, something not yet done or checked, or a word on what you asked for; read on the whole message, past the `label`'s clip, and used by the final panel's balloon), `question` and, for prompts, `origin`. The advisor (a server tool) is an event with tool `Advisor` and intent `helper`; its answer is encrypted in the transcript. In plan mode the plan file under `~/.claude/plans/` is planning (intent `plan`, its numbered steps as `todos`), and `ExitPlanMode`'s result carries the plan (`output`) and your answer (`Approved` / `Sent back`).
-- `panels`: at most 12, each `{ kind, events, caption: { story, director }, balloon?, files?, notes?, helpers?, todos?, tests?, ok?, rounds?, roundsOk?, question?, live? }`, with `events` as indices into `events`. `balloon.who` is `you`, `claude` or `note` (words that start a turn but are not yours: a background job's report, another session's message). A loop's `roundsOk` lists every test run in order, `true` where it passed and `null` for one still running, so a failure inside a round that passed in the end still shows, and a live loop shows the runs that finished. Kinds: `mission`, `investigate`, `research`, `plan`, `backup`, `fix`, `check`, `loop` (fix and check rounds folded), `run`, `save`, `ask`, `compact`, `montage` (quieter panels folded to fit) and `complete`.
+- `panels`: at most 12, each `{ kind, events, caption: { story, director }, balloon?, files?, notes?, helpers?, todos?, tests?, ok?, rounds?, roundsOk?, question?, live? }`, with `events` as indices into `events`. A run panel's `printed` holds the first lines its last command printed (its last lines when it failed), for the terminal card: the card is drawn on the canvas and goes into a saved image, so lines that look like secrets are left out and addresses and long token-like strings are masked. `balloon.who` is `you`, `claude` or `note` (words that start a turn but are not yours: a background job's report, another session's message). A loop's `roundsOk` lists every test run in order, `true` where it passed and `null` for one still running, so a failure inside a round that passed in the end still shows, and a live loop shows the runs that finished. Kinds: `mission`, `investigate`, `research`, `plan`, `backup`, `fix`, `check`, `loop` (fix and check rounds folded), `run`, `save`, `ask`, `compact`, `montage` (quieter panels folded to fit) and `complete`.
 - `filesChanged`, `added`, `removed` (summed per edit), `helpers` (the turn's own helper calls; a report turn's prompt carries the helper it answers, which is not counted), `helpersRunning`, `noneWorked` (every tool call of the turn failed, so the last panel is no celebration), `durationMs`, `partial` (the turn began before the part that was read) and `truncated` (older outputs, hunks and helper texts were dropped to keep the page small: a page carries at most 120,000 characters of them, newest first; an edit that lost its lines keeps its counts and an empty `hunks`).
 
 ## Git worktrees of a case

src/story-mode.ts

@@ -2164,6 +2164,19 @@ function montageSummary(evs: StoryEvent[]): string {
   return parts.length ? parts.join(', ') : 'a few more steps';
 }
 
+/**
+ * One output line as the terminal card may show it. The card is drawn on the canvas, so it goes into a
+ * saved image someone may share: a line that looks like a secret is left out, addresses and long
+ * token-like strings are masked, and Claude Code's own markers go.
+ */
+function printableLine(line: string): string | null {
+  const text = line.replace(/<\/?[a-z_]+>/g, '').trim();
+  if (!text || /^\(Bash completed with no output\)$/.test(text)) return null;
+  // Anywhere in the line, inside identifiers too (`API_TOKEN=`, `db_password:`): losing a harmless line is cheap.
+  if (/token|secret|passw(or)?d|api[_-]?key|authorization|bearer|private[_ -]?key|credential/i.test(text)) return null;
+  return clip(text.replace(/[\w.+-]+@[\w-]+(\.[\w-]+)+/g, '…@…').replace(/[A-Za-z0-9+/=_-]{32,}/g, '…'), 90);
+}
+
 /** 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;
@@ -2439,8 +2452,18 @@ function decorate(draft: DraftPanel, events: StoryEvent[], turn: StoryTurn, inde
       }
       break;
     }
-    case 'run':
+    case 'run': {
       panel.notes = notesOf(evs);
+      // What it printed, so a dry run or a listing you asked to see is on the page, not only in the drawer.
+      const lastOut = [...evs].reverse().find((e) => e.output?.trim() && !e.running);
+      if (lastOut?.output) {
+        const lines = lastOut.output
+          .split('\n')
+          .map(printableLine)
+          .filter((l): l is string => !!l);
+        const picked = lastOut.ok === false ? lines.slice(-2) : lines.slice(0, 3);
+        if (picked.length) panel.printed = picked;
+      }
       story = evs.every((e) => e.running)
         ? `Claude starts ${evs.length === 1 ? 'a command' : `${evs.length} commands`} in the background.`
         : errors && errors === evs.length
@@ -2456,6 +2479,7 @@ function decorate(draft: DraftPanel, events: StoryEvent[], turn: StoryTurn, inde
             : 'Claude works in the terminal.';
       director = `${tools} · ${shortCommand(evs[0])}${errors ? ` · ${plural(errors, 'error')}` : ''}`;
       break;
+    }
     case 'save': {
       panel.notes = notesOf(evs);
       const last = evs[evs.length - 1];

src/types/story.ts

@@ -167,6 +167,8 @@ export interface StoryPanel {
   balloon?: { who: 'you' | 'claude' | 'note'; text: string };
   files?: string[];
   notes?: string[];
+  /** What a run panel's last command printed, for its terminal card: the first lines, the last ones when it failed. */
+  printed?: string[];
   helpers?: Array<{ name: string; type: string; task?: string; event: number }>;
   added?: number;
   removed?: number;

src/web/public/story-mode.js

@@ -688,7 +688,8 @@ const StoryArt = (() => {
     }
     return { w, h };
   }
-  function terminal(ctx, x, y, w, h, lines, f) {
+  /** A terminal card: the commands (`› …`), then, dimmer, the first lines one of them printed. */
+  function terminal(ctx, x, y, w, h, lines, f, printed = []) {
     rr(ctx, x, y, w, h, 7);
     inked(ctx, C.line, 2.5);
     ctx.fillStyle = C.balloonInk;
@@ -708,10 +709,21 @@ const StoryArt = (() => {
       ctx.fillStyle = i === Math.min(rows, lines.length) - 1 ? C.yellow : C.balloon;
       ctx.fillText(text.slice(0, shown), x + 9, y + 30 + i * 14);
     });
-    if (f % 30 < 15) {
-      const n = Math.min(rows, lines.length);
+    const n = Math.min(rows, lines.length);
+    const out = printed.slice(0, Math.max(0, rows - n));
+    out.forEach((line, i) => {
+      const text = fitText(ctx, line, w - 18);
+      const at = 8 + (n + i) * 12;
+      const shown = Math.floor(interpolate(f, [at, at + 10], [0, text.length]));
+      ctx.save();
+      ctx.globalAlpha = 0.72;
       ctx.fillStyle = C.balloon;
-      ctx.fillRect(x + 9, y + 22 + n * 14, 6, 10);
+      ctx.fillText(text.slice(0, shown), x + 9, y + 30 + (n + i) * 14);
+      ctx.restore();
+    });
+    if (f % 30 < 15) {
+      ctx.fillStyle = C.balloon;
+      ctx.fillRect(x + 9, y + 22 + (n + out.length) * 14, 6, 10);
     }
   }
   function globe(ctx, cx, cy, r, f) {
@@ -1248,7 +1260,7 @@ const StoryArt = (() => {
       draw(ctx, W, H, f, p, hit) {
         const tw = Math.min(W * 0.62, 260);
         const notes = p.panel.notes?.length ? p.panel.notes : ['…'];
-        terminal(ctx, 14, 92, tw, 140, notes, f);
+        terminal(ctx, 14, 92, tw, 140, notes.slice(0, p.panel.printed?.length ? 3 : 8), f, p.panel.printed || []);
         hit('output', 14, 92, tw, 140, 'See the commands and their output');
         agent(ctx, W - 48, 250, {
           frame: f,

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

@@ -22,6 +22,7 @@
         "balloon": "Build a tiny Python word-count tool here (no git, no installs). First, use the Agent tool TWICE with run_in_background: true to start two background helper agents in parallel: helper A writes README.md describing wordcount.py (usage: python3 wordcount.py FILE prints lines, words, chars), helper B writes extra pytest tests in tests/test_extra.py for count(text) returning a (lines, words, chars) tu…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": "Starting both helpers in parallel first.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -55,6 +58,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -65,6 +69,7 @@
         "balloon": "I wrote wordcount.py and tests/test_basic.py, and the one python3 -m pytest -q run passed 4 of 4 tests. Only 4 tests ran, so tests/test_extra.py from helper B wasn't there yet and those tests haven't been run.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -92,6 +97,7 @@
         "balloon": "Agent \"Helper A: write README.md\" finished",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -102,6 +108,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -112,6 +119,7 @@
         "balloon": "Helper A has finished writing README.md, and I checked it against the code. It describes the command-line usage, count(text) and how to run the tests correctly, and its example ((2, 4, 24)) is right. It doesn't mention that running the tool without exactly one file prints a usage message and exits…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -139,6 +147,7 @@
         "balloon": "Agent \"Helper B: write extra pytest tests\" finished",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -152,6 +161,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -162,6 +172,7 @@
         "balloon": "Both helpers have finished.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -189,6 +200,7 @@
         "balloon": "Follow-up, no git: first start ONE background helper with the Agent tool (run_in_background: true) that writes CHANGELOG.md slowly: it must run sleep 20 with Bash, then Write CHANGELOG.md with one entry, then run sleep 20 again, then add a second entry with the Edit tool. Next, start the full test suite in the background with the Bash tool's run_in_background: true parameter, as: sleep 30; python…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -199,6 +211,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -209,6 +222,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -219,6 +233,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -229,6 +244,7 @@
         "balloon": "I added the --json flag and started the helper and the test run as you asked. The helper and the tests are still running in the background; I haven't waited for either, so I don't have results yet. No git, as you asked.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -256,6 +272,7 @@
         "balloon": "Background command \"Run full test suite after a 30s delay\" completed (exit code 0)",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -269,6 +286,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -279,6 +297,7 @@
         "balloon": "The background test run finished and all 23 tests passed, including the new --json test.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -306,6 +325,7 @@
         "balloon": "Agent \"Helper: write CHANGELOG.md slowly\" finished",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -316,6 +336,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -326,6 +347,7 @@
         "balloon": "Everything you asked for this round is done. The helper used four tools, one for each step you asked for, so it looks like it followed the sleep, write, sleep, edit sequence.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -353,6 +375,7 @@
         "balloon": "Small follow-up, no git: run python3 -m pytest -q once, then show the first 5 lines of CHANGELOG.md with head -5, then reply with one sentence.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -366,6 +389,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -376,6 +400,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -386,6 +411,7 @@
         "balloon": "All 23 tests passed, and the first five lines of CHANGELOG.md are the # Changelog title, the ## 2026-10-18 heading and the wordcount.py entry, with the --json entry on line 6, just past what head -5 shows.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -413,6 +439,7 @@
         "balloon": "Reply with just the word OK, no tools.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -423,6 +450,7 @@
         "balloon": "OK",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -450,6 +478,7 @@
         "balloon": "Reply with just the word DONE, no tools.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -460,6 +489,7 @@
         "balloon": "DONE",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/fixtures/story-corpus/bash-script-background-helper.expected.json

@@ -22,6 +22,7 @@
         "balloon": "First write a short todo list for this job, then build a bash script rename_by_date.sh in this folder that renames *.jpg files to their modification time as YYYY-MM-DD_HHMMSS.jpg with a --dry-run flag, and while you do that use the Agent tool to have one helper agent write README.md for the script; finally test it on a few sample files you create in a samples folder.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -62,6 +66,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -72,6 +77,7 @@
         "balloon": "Now I'll create the samples and run the dry run.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 5
       },
       {
@@ -82,6 +88,11 @@
         "balloon": "My mk -- ... helper call left a stray file named --. I'll remove it, then run the dry run and save a listing to confirm nothing changed.",
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "--bogus exit=2",
+          "two dirs exit=2",
+          "-- two dirs exit=2"
+        ],
         "events": 5
       },
       {
@@ -92,6 +103,7 @@
         "balloon": "All the script tests pass.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -119,6 +131,7 @@
         "balloon": "Agent \"Write README for rename script\" finished",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -129,6 +142,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -139,6 +153,11 @@
         "balloon": "The README matches the script.",
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "mv: cannot move '….jpg' to '….jpg': Permission denied",
+          "exit=1",
+          "q.jpg"
+        ],
         "events": 1
       },
       {
@@ -149,6 +168,7 @@
         "balloon": "I built rename_by_date.sh, the helper agent wrote README.md, and the script passed every test I ran on sample files in samples/. Nothing is committed because this folder isn't a git repository, even though the session context said it was. Only lowercase .jpg is matched, as you asked.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -176,6 +196,7 @@
         "balloon": "Add a --prefix option to rename_by_date.sh that puts a given text in front of the new names, and show me a dry run on the samples.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -186,6 +207,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -196,6 +218,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -206,6 +229,11 @@
         "balloon": "Here's the dry run you asked for on the samples, plus a check that nothing on disk changed:",
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "--- missing value",
+          "exit=2 error: --prefix needs a value",
+          "--- slash"
+        ],
         "events": 2
       },
       {
@@ -216,6 +244,7 @@
         "balloon": "All passing.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -226,6 +255,7 @@
         "balloon": "I added a --prefix option to rename_by_date.sh. The samples folder didn't change.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/fixtures/story-corpus/compact-then-plan-mode.expected.json

@@ -22,6 +22,7 @@
         "balloon": "Write mathx.py with a function clamp(value, low, high) that returns value limited to the range low..high and raises ValueError when low > high, then write test_mathx.py with exactly 5 pytest tests for it, and run them with python3 -m pytest -q.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -65,6 +69,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -75,6 +80,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -85,6 +91,7 @@
         "balloon": "I wrote clamp and its five tests, and all five pass. I didn't commit CLAUDE.md, .claude/ or .codeman-agent-case.json.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -112,6 +119,7 @@
         "balloon": "/compact",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -122,6 +130,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -132,6 +141,7 @@
         "balloon": "Compacted",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -159,6 +169,7 @@
         "balloon": "Call your EnterPlanMode tool to enter plan mode, then write a short 3-step plan for adding a wrap(value, low, high) function (modular wrap into the range) to mathx.py with 3 more pytest tests, and present that plan with your ExitPlanMode tool. After I approve, implement it and run the tests.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -169,6 +180,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -179,6 +191,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -189,6 +202,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -199,6 +213,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -209,6 +224,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -219,6 +235,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -232,6 +249,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -242,6 +260,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -252,6 +271,7 @@
         "balloon": "I added wrap() as planned, with 3 new tests.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

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

@@ -22,6 +22,7 @@
         "balloon": "Build a tiny Python module calc_stats.py with mean(xs) and median(xs), strictly test-first: FIRST write tests/test_calc_stats.py that imports calc_stats, THEN run python3 -m pytest -q once while calc_stats.py does not exist yet (it must error at collection), THEN write calc_stats.py and run python3 -m pytest -q again until it passes. Install nothing, stay in this folder.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -62,6 +66,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -76,6 +81,7 @@
           "errors": 1
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -86,6 +92,7 @@
         "balloon": "Red as expected: collection error, exit code 2.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -96,6 +103,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -109,6 +117,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -119,6 +128,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -129,6 +139,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -139,6 +150,7 @@
         "balloon": "I built it test-first: the suite failed at collection before calc_stats.py existed, then passed once the module was written. I didn't push, didn't touch origin.git/ and didn't create a README, since none existed to keep in sync.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -166,6 +178,7 @@
         "balloon": "Add mode(xs) to calc_stats.py (ValueError on empty input, ties return the smallest value) in two separate steps without committing in between and without consulting the advisor: first add mode tests to tests/test_calc_stats.py and run python3 -m pytest -q (it should fail because mode is missing), then implement mode in calc_stats.py and run python3 -m pytest -q again until green; only then git co…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -182,6 +195,7 @@
           false,
           true
         ],
+        "printed": null,
         "events": 5
       },
       {
@@ -192,6 +206,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -202,6 +217,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -212,6 +228,7 @@
         "balloon": "mode(xs) is in calc_stats.py, committed once after the tests went green, and pushed. I didn't use the advisor, gh or the network.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -239,6 +256,7 @@
         "balloon": "In ONE Bash command, append exactly two lines to NOTES.md with cat >> NOTES.md <<'EOF' (two short lines about mode) EOF and then cat calc_stats.py; next run python3 -m pytest -q; then git add NOTES.md and commit it, and push with exactly this command, run as written with nothing appended: git push && git describe --tags (it is expected to fail after pushing; just report it).",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -249,6 +267,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -262,6 +281,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -272,6 +292,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -282,6 +303,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -292,6 +314,7 @@
         "balloon": "NOTES.md has the two lines and is committed and pushed. As you expected, git describe --tags then failed because the repo has no tags.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -319,6 +342,7 @@
         "balloon": "Reply with just the word ok, use no tools.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -329,6 +353,7 @@
         "balloon": "ok",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

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

@@ -22,6 +22,7 @@
         "balloon": "Create a small JavaScript ES module slugify.mjs exporting slugify(str) (lowercase, trim, spaces and punctuation to single hyphens, no leading or trailing hyphens) plus slugify.test.mjs with 4 node:test tests; on purpose make ONE test expectation wrong so the first run of node --test fails, run it, then fix that expectation, rerun until all pass, then git init if needed and git commit both files w…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -58,6 +61,7 @@
           false,
           true
         ],
+        "printed": null,
         "events": 5
       },
       {
@@ -68,6 +72,10 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "Initialized empty Git repository in /work/cases/story-r3-c/.git/",
+          "name= email=…@…"
+        ],
         "events": 1
       },
       {
@@ -78,6 +86,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -88,6 +97,7 @@
         "balloon": "slugify.mjs and its test file are written, all 4 tests pass, and both files are committed.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -115,6 +125,7 @@
         "balloon": "Add a fifth node:test test for slugify('') to slugify.test.mjs that on purpose expects 'x' so the first run fails, run node --test, then fix that expectation to the empty string, rerun until all pass, and commit the change.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -131,6 +142,7 @@
           false,
           true
         ],
+        "printed": null,
         "events": 4
       },
       {
@@ -141,6 +153,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -151,6 +164,7 @@
         "balloon": "I added a fifth test for slugify('').",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

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

@@ -22,6 +22,7 @@
         "balloon": "Build a small Python package inventory/ (a CLI to add, list and remove items kept in a JSON file) split into at least 4 modules (for example storage.py, models.py, commands.py, cli.py), with a pytest suite of at least 25 tests in total; build it one module at a time, and after EACH module write its tests and run pytest, fixing any failures before moving on to the next module; then git init the fo…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 4
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -62,6 +66,7 @@
         "balloon": "Also add a --csv option to the list command, with tests.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -81,6 +86,7 @@
           false,
           true
         ],
+        "printed": null,
         "events": 14
       },
       {
@@ -91,6 +97,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -101,6 +108,11 @@
         "balloon": "README examples match the real output.",
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "./CLAUDE.md",
+          "./.claude/settings.local.json",
+          "./.codeman-agent-case.json"
+        ],
         "events": 2
       },
       {
@@ -111,6 +123,7 @@
         "balloon": "Commits are in.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -124,6 +137,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -134,6 +148,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -144,6 +159,7 @@
         "balloon": "The inventory/ package is built and committed. --csv is part of the CLI commit rather than its own commit, because you asked for it before that module existed.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -171,6 +187,7 @@
         "balloon": "Now do these 15 small steps strictly one at a time, each as its own separate tool call, in exactly this order, never combining two steps into one call and without extra steps in between: 1) Read inventory/models.py with the Read tool; 2) with the Edit tool add the line 'Version 0.1.0.' at the end of README.md; 3) run 'python3 -m pytest -q' in Bash; 4) run 'uname -s' in Bash; 5) run 'git add -A &&…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -181,6 +198,7 @@
         "balloon": "Step 1:",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 6
       },
       {
@@ -191,6 +209,7 @@
         "balloon": "Also, as a 16th step at the very end, run git log --oneline in Bash.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -201,6 +220,7 @@
         "balloon": "Noted, I'll add step 16 at the end.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -211,6 +231,7 @@
         "balloon": "Step 10:",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -221,6 +242,7 @@
         "balloon": "Step 11:",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -231,6 +253,7 @@
         "balloon": "Step 12:",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -244,6 +267,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -254,6 +278,9 @@
         "balloon": "Step 14:",
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "x86_64"
+        ],
         "events": 1
       },
       {
@@ -264,6 +291,7 @@
         "balloon": "Step 15:",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -274,6 +302,7 @@
         "balloon": "Step 16:",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -284,6 +313,7 @@
         "balloon": "I ran all 16 steps in order, one tool call each. I didn't re-run ruff, because only README.md and CHANGELOG.md changed, and ruff doesn't check Markdown.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

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

@@ -22,6 +22,7 @@
         "balloon": "Build a small Python command-line tool wcount.py in this folder that prints the line, word and character counts of the files given as arguments, like wc, with a --json flag that prints the counts as JSON, plus pytest tests in test_wcount.py; run pytest and fix anything that fails until all tests pass.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -59,6 +62,7 @@
           false,
           true
         ],
+        "printed": null,
         "events": 7
       },
       {
@@ -69,6 +73,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -79,6 +84,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -89,6 +95,11 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "wcount: nope: No such file or directory",
+          "wcount: nope2: No such file or directory",
+          "0 0 0 total"
+        ],
         "events": 1
       },
       {
@@ -99,6 +110,7 @@
         "balloon": "I built wcount.py and test_wcount.py. With a missing file plus a readable one, the total row didn't appear because the code counted files that could be read rather than files that were named.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/fixtures/story-corpus/question-tests-first-push.expected.json

@@ -22,6 +22,7 @@
         "balloon": "Build a tiny Python command-line tool in this folder that turns text into URL slugs. Before doing anything else, ask me with your AskUserQuestion tool whether the tool should be named slugify or kebab (offer exactly those two options). After my answer, first write me your plan as a short numbered list, then write the pytest tests FIRST (at least 8 tests in tests/), run them with python3 -m pytest…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -68,6 +72,7 @@
           false,
           true
         ],
+        "printed": null,
         "events": 6
       },
       {
@@ -78,6 +83,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -88,6 +94,11 @@
         "balloon": "Now I'll run each documented README command by hand and check the git identity:",
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "Initialized empty Git repository in /work/cases/story-r2-c/.git/",
+          "A  .gitignore",
+          "A  CLAUDE.md"
+        ],
         "events": 3
       },
       {
@@ -98,6 +109,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -111,6 +123,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -121,6 +134,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -131,6 +145,7 @@
         "balloon": "I've built the kebab slug tool in this folder, committed it, and pushed it to the new bare repo remote.git. Not tested: the README says pip install . gives you a kebab command.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/fixtures/story-corpus/teammates-then-interrupt.expected.json

@@ -22,6 +22,7 @@
         "balloon": "Create a tiny node project here: package.json (type module) and index.js exporting two functions add(a,b) and slugify(s). Then start TWO helpers with the Agent tool IN THE BACKGROUND (run_in_background: true): one writes README.md describing the project, the other writes CONTRIBUTING.md. While they work, write index.test.mjs using node:test with a few tests for both functions and run it with node…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -62,6 +66,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -75,6 +80,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -85,6 +91,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -95,6 +102,7 @@
         "balloon": "I've set up the project, started both helpers, and the tests pass: node --test index.test.mjs ran 8 tests, 8 passed, 0 failed. No linter is set up in this project, so I didn't run one.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -122,6 +130,7 @@
         "balloon": "contributing-writer: I wrote /work/cases/story-r2-b/CONTRIBUTING.md covering setup, tests, code style, commits, README sync and PRs; it isn't committed because git was off-limits, and index.test.mjs and README.md, which…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -132,6 +141,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -142,6 +152,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -152,6 +163,7 @@
         "balloon": "The CONTRIBUTING.md helper has finished, and I've committed its file (docs: add CONTRIBUTING.md). The helper reported that index.test.mjs and README.md didn't exist yet. That's out of date: it must have looked before I wrote the tests, and both files are there now. README.md is already on disk, but its helper hasn't reported back yet, so I haven't committed it in case it's still being written.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -179,6 +191,7 @@
         "balloon": "readme-writer: README done, test file missing",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -189,6 +202,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -202,6 +216,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -212,6 +227,7 @@
         "balloon": "The project is done: both helpers finished, and I checked their files and committed them. Both helpers warned that index.test.mjs didn't exist, but they had looked before I created it, so the README's Testing section is correct.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -239,6 +255,7 @@
         "balloon": "Big refactor: split index.js into three modules under lib/ (lib/add.js, lib/slugify.js, lib/util.js with shared helpers), keep index.js as a re-export barrel, add full JSDoc to every function, add at least 15 more node:test tests in separate files under test/, update README and CONTRIBUTING to match, run the tests after each step and commit each step separately.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -249,6 +266,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -262,6 +280,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -272,6 +291,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -285,6 +305,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -295,6 +316,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -305,6 +327,7 @@
         "balloon": "Step 3: new tests under test/.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -332,6 +355,7 @@
         "balloon": "Probe turn, no other work: run the shell command sleep 65 with Bash (foreground, not background), then reply with just the word done.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -342,6 +366,9 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "Blocked: standalone sleep 65. To wait for a condition, use Monitor with an until-loop (e.…"
+        ],
         "events": 1
       },
       {
@@ -352,6 +379,7 @@
         "balloon": "Not done. The harness blocked sleep 65: a sleep on its own in the foreground isn't allowed. I didn't try to get around the block, for example by chaining shorter sleeps or running it in the background.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/fixtures/story-corpus/text-plan-dry-run-helper.expected.json

@@ -22,6 +22,7 @@
         "balloon": "First reply with a numbered plan of exactly 4 steps (as text, not a tool). Then build csvstat.py, a Python CLI that prints row count and per-column min/max/mean of a CSV, with pytest tests. Do a dry run on a sample CSV and save its log with `| tee /tmp/csvstat-dryrun.log`. Start ONE helper with the Agent tool, run_in_background: true and WITHOUT a name parameter, to write README.md; do not wait f…",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -42,6 +44,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -52,6 +55,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -62,6 +66,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 4
       },
       {
@@ -75,6 +80,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -85,6 +91,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -95,6 +102,11 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "rows: 5",
+          "column      min     max     mean",
+          "city        -       -       -"
+        ],
         "events": 1
       },
       {
@@ -105,6 +117,7 @@
         "balloon": "I've built csvstat.py, a command-line tool that prints a CSV's row count and the min, max and mean of each column. A background helper is writing README.md; I didn't wait for it, so it isn't confirmed yet. The population mean shows a problem, described below.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]
@@ -132,6 +145,7 @@
         "balloon": "Agent \"Write README for csvstat\" finished",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -142,6 +156,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -152,6 +167,7 @@
         "balloon": "The background helper has finished README.md, and I've checked its work. Rounding: the README now explains that numbers are shown to 6 significant digits, so the population mean of 214574.2 appears as 214574. That note makes the current behaviour clear, but it doesn't fix it.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/fixtures/story-corpus/web-page-node-test.expected.json

@@ -22,6 +22,7 @@
         "balloon": "Build a tiny single-page pomodoro timer in this folder: index.html with inline CSS, timer.js exporting the timer logic as plain functions, and timer.test.mjs using node:test that covers start, pause and reset; run the tests with node --test until they pass, then git init and commit everything with a short message, with no remote and no push.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -32,6 +33,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 2
       },
       {
@@ -42,6 +44,10 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": [
+          "['hooks'] {'hooks': ['Notification', 'Stop', 'UserPromptSubmit', 'SubagentStop', 'Teammat…",
+          "…@…"
+        ],
         "events": 1
       },
       {
@@ -52,6 +58,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -62,6 +69,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 3
       },
       {
@@ -75,6 +83,7 @@
           "failed": 0
         },
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -85,6 +94,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -95,6 +105,7 @@
         "balloon": null,
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 1
       },
       {
@@ -105,6 +116,7 @@
         "balloon": "I built the pomodoro timer. Opening the page: it won't work if you double-click index.html. The page loads timer.js as a module, and Chrome and Firefox block that for local files, so you'd get a timer that never moves.",
         "tests": null,
         "roundsOk": null,
+        "printed": null,
         "events": 0
       }
     ]

test/story-corpus.test.ts

@@ -49,6 +49,7 @@ function pagesOf(name: string) {
         balloon: x.balloon?.text ?? null,
         tests: x.tests ?? null,
         roundsOk: x.roundsOk ?? null,
+        printed: x.printed ?? null,
         events: x.events.length,
       })),
     };

test/story-mode.test.ts

@@ -1034,6 +1034,24 @@ describe('buildStoryPage', () => {
     );
   });
 
+  it("a run panel's terminal card shows what the command printed, never what looks like a secret", () => {
+    const page = pageOf([
+      prompt('dry run please'),
+      ...bash(
+        'b1',
+        './rename.sh --dry-run samples/',
+        'Dry run',
+        'a.jpg -> holiday_2024-05-01.jpg\nAPI_TOKEN=abc123\nauthor: me@example.com\nb.jpg -> holiday_2024-05-02.jpg'
+      ),
+      end(1000),
+    ]);
+    expect(page.panels.find((p) => p.kind === 'run')!.printed).toEqual([
+      'a.jpg -> holiday_2024-05-01.jpg',
+      'author: …@…',
+      'b.jpg -> holiday_2024-05-02.jpg',
+    ]);
+  });
+
   it('a helper messaged twice reads as one finished run, not as superseded', () => {
     const r = createStoryReducer({ sidechain: true });
     const lines = [prompt('write it'), said('Done.'), prompt('also the tests'), said('Done too.')].map((l) => ({