@bendyline/gilde 0.1.51 → 0.1.53

package/data/craftbook-templates/fl/flaky-test-fix/versions/1.0.1/test.json

@@ -0,0 +1,239 @@
+{
+  "schemaVersion": 1,
+  "title": "Deflake an order-dependent test with receipts",
+  "objective": "The craftbook drives a real deflake of a seeded order-dependence flake: statistical reproduction with recorded failing runs, a mechanism-level diagnosis, cleanup at the level every test shares (never skipping or weakening the test), and a stability proof - verified mechanically, including a five-run consecutive-green oracle, a late-probe order-robustness check, and a mutant restoring the original leaking order.",
+  "tags": [
+    "code",
+    "flaky-test",
+    "tests",
+    "tactical-fleet"
+  ],
+  "prompt": "Use the Fix a Flaky Test craftbook: our suite is red in CI but the failing test passes when people run it by itself, so it keeps getting waved through. The report is in source/flake-report.md. Find the nondeterminism, remove it for real, and prove the suite stays green.",
+  "setup": {
+    "projectName": "Flagship Flags",
+    "about": "A tiny feature-flags module and its test suite. Tests run with `npm run test` (node --test).",
+    "files": [
+      {
+        "path": "package.json",
+        "content": "{\n  \"name\": \"flagship\",\n  \"private\": true,\n  \"version\": \"1.0.0\",\n  \"type\": \"module\",\n  \"scripts\": {\n    \"test\": \"node --test\"\n  }\n}\n",
+        "modelInput": false
+      },
+      {
+        "path": "src/flags.js",
+        "content": "const flags = new Map();\n\nexport function enable(name) {\n  flags.set(name, true);\n}\n\nexport function disable(name) {\n  flags.delete(name);\n}\n\nexport function isEnabled(name) {\n  return flags.get(name) === true;\n}\n\nexport function enabledCount() {\n  return flags.size;\n}\n\nexport function resetFlags() {\n  flags.clear();\n}\n"
+      },
+      {
+        "path": "tests/flags.test.mjs",
+        "content": "import assert from 'node:assert/strict';\nimport { test } from 'node:test';\nimport { enable, disable, isEnabled, enabledCount } from '../src/flags.js';\n\ntest('enable turns a flag on', () => {\n  enable('beta-banner');\n  assert.equal(isEnabled('beta-banner'), true);\n});\n\ntest('disable turns a flag off again', () => {\n  enable('holiday-theme');\n  disable('holiday-theme');\n  assert.equal(isEnabled('holiday-theme'), false);\n});\n\ntest('unknown flags read as disabled', () => {\n  assert.equal(isEnabled('dark-mode'), false);\n});\n\ntest('a fresh session starts with no flags enabled', () => {\n  assert.equal(enabledCount(), 0);\n});\n"
+      },
+      {
+        "path": "source/flake-report.md",
+        "content": "# Flake report\n\nThe suite has been red in CI all week, and every developer who ran the failing test by itself saw it pass - so it kept getting waved through.\n\n    npm run test\n    not ok 4 - a fresh session starts with no flags enabled\n      AssertionError: 1 !== 0\n\nRunning the full suite fails on that test consistently; running just that one test with `--test-name-pattern \"fresh session\"` is green every time. Nobody has touched `src/flags.js` in weeks.\n\nConstraints:\n\n- The pristine-session test guards a real invariant (feature sessions must start clean). Do not delete or weaken it.\n- The app depends on `src/flags.js` exactly as it is - fix the tests' hygiene, not the module's API.\n",
+        "modelInput": true
+      },
+      {
+        "path": "tests/verify-deflake.mjs",
+        "content": "import assert from 'node:assert/strict';\nimport { execFileSync } from 'node:child_process';\nimport { cpSync, mkdirSync, readFileSync, readdirSync, rmSync, writeFileSync } from 'node:fs';\nimport { join } from 'node:path';\nimport { enable, isEnabled, resetFlags, enabledCount } from '../src/flags.js';\n\nconst root = process.cwd();\nconst ORACLE_FILE = 'verify-deflake.mjs';\n\n// 1) The flags module still means what it meant.\nresetFlags();\nenable('oracle-probe');\nassert.equal(isEnabled('oracle-probe'), true, 'enable must still work');\nassert.equal(enabledCount(), 1, 'enabledCount must still count');\nresetFlags();\nassert.equal(enabledCount(), 0, 'resetFlags must still clear');\nconsole.log('DEFLAKE_ORACLE flags semantics ok');\n\nfunction runSuite(cwd) {\n  try {\n    execFileSync(process.execPath, ['--test'], { cwd, stdio: 'pipe', timeout: 30000 });\n    return 0;\n  } catch (err) {\n    return typeof err.status === 'number' ? err.status : 1;\n  }\n}\n\nfunction copyTree(dest) {\n  rmSync(dest, { recursive: true, force: true });\n  mkdirSync(dest, { recursive: true });\n  cpSync(join(root, 'src'), join(dest, 'src'), { recursive: true });\n  cpSync(join(root, 'tests'), join(dest, 'tests'), { recursive: true });\n  rmSync(join(dest, 'tests', ORACLE_FILE), { force: true });\n  cpSync(join(root, 'package.json'), join(dest, 'package.json'));\n  return dest;\n}\n\n// 2) Five consecutive green runs of the real suite.\nfor (let run = 1; run <= 5; run += 1) {\n  assert.equal(runSuite(root), 0, `full-suite run ${run} of 5 must be green`);\n}\nconsole.log('DEFLAKE_ORACLE stable 5x');\n\n// 3) The pristine-session assertion survives, unweakened and unskipped.\nconst testFiles = readdirSync(join(root, 'tests')).filter(\n  (name) => name !== ORACLE_FILE && /\\.(mjs|cjs|js)$/.test(name),\n);\nconst combined = testFiles\n  .map((name) => readFileSync(join(root, 'tests', name), 'utf8'))\n  .join('\\n');\nassert.match(\n  combined,\n  /enabledCount\\s*\\(\\s*\\)\\s*(?:,\\s*0|===\\s*0)/,\n  'the pristine-session assertion (enabledCount() must be 0) has to survive',\n);\nassert.ok(\n  !/\\.skip\\s*\\(|\\.todo\\s*\\(|\\.only\\s*\\(|\\{\\s*skip\\s*:|\\{\\s*todo\\s*:/.test(combined),\n  'no test may be skipped, stubbed as todo, or narrowed with only',\n);\nconsole.log('DEFLAKE_ORACLE assertion intact');\n\n// 4) Order robustness: tests appended at the END of the suite must inherit\n// the cleanup. The first probe enables a flag and never cleans up; the\n// second expects a pristine session anyway. Only hygiene that covers every\n// test (beforeEach/afterEach-style) passes both - a one-off reset inside\n// the seeded failing test, or a reorder, fails here. Aliased imports keep\n// the probe valid whatever the model did to the file's own imports.\nconst hostFile = testFiles.find((name) =>\n  /enabledCount/.test(readFileSync(join(root, 'tests', name), 'utf8')),\n);\nassert.ok(hostFile, 'no test file references enabledCount');\nconst probeRoot = copyTree(join(root, '.oracle-probe'));\nconst probeTests =\n  readFileSync(join(probeRoot, 'tests', hostFile), 'utf8') +\n  \"\\n\\nimport { test as zzTest } from 'node:test';\\nimport zzAssert from 'node:assert/strict';\\nimport { enable as zzEnable, isEnabled as zzIsEnabled, enabledCount as zzEnabledCount } from '../src/flags.js';\\n\\nzzTest('zz probe enables a flag of its own', () => {\\n  zzEnable('zz-probe-flag');\\n  zzAssert.equal(zzIsEnabled('zz-probe-flag'), true);\\n});\\n\\nzzTest('zz probe still gets a pristine session', () => {\\n  zzAssert.equal(zzEnabledCount(), 0);\\n});\\n\";\nwriteFileSync(join(probeRoot, 'tests', hostFile), probeTests);\nassert.equal(\n  runSuite(probeRoot),\n  0,\n  'a test appended after the suite must still see a pristine session - cleanup has to cover every test, not only the seeded order',\n);\nconsole.log('DEFLAKE_ORACLE order robust');\n\n// 5) Mutant: restore the ORIGINAL leaking test file - the suite must fail\n// again, proving the seeded defect is still detectable and the green suite\n// comes from removed nondeterminism, not from a neutered module or dodge.\nconst ORIGINAL_TEST_FILE = `import assert from 'node:assert/strict';\nimport { test } from 'node:test';\nimport { enable, disable, isEnabled, enabledCount } from '../src/flags.js';\n\ntest('enable turns a flag on', () => {\n  enable('beta-banner');\n  assert.equal(isEnabled('beta-banner'), true);\n});\n\ntest('disable turns a flag off again', () => {\n  enable('holiday-theme');\n  disable('holiday-theme');\n  assert.equal(isEnabled('holiday-theme'), false);\n});\n\ntest('unknown flags read as disabled', () => {\n  assert.equal(isEnabled('dark-mode'), false);\n});\n\ntest('a fresh session starts with no flags enabled', () => {\n  assert.equal(enabledCount(), 0);\n});\n`;\nconst mutantRoot = copyTree(join(root, '.oracle-mutant'));\nwriteFileSync(join(mutantRoot, 'tests', 'flags.test.mjs'), ORIGINAL_TEST_FILE);\nassert.notEqual(\n  runSuite(mutantRoot),\n  0,\n  'restoring the original leaking order must make the suite fail again - the seeded defect stopped being detectable',\n);\nconsole.log('DEFLAKE_ORACLE mutant caught');\nconsole.log('DEFLAKE_ORACLE done');\n",
+        "surface": "harness"
+      }
+    ],
+    "craftbookParams": {
+      "scope": "The test 'a fresh session starts with no flags enabled' in tests/flags.test.mjs fails whenever the whole suite runs and passes by itself; see source/flake-report.md. Make the suite deterministic without weakening that test."
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "The order-dependence mechanism is removed with hygiene that covers every test, the pristine-session assertion survives unweakened, five consecutive suite runs are green, a late-appended probe still sees a pristine session, and restoring the original test file makes the suite fail again.",
+    "deliverables": [
+      {
+        "path": "{{task.dir}}/flake-evidence.md",
+        "kind": "markdown-notes",
+        "artifact": true,
+        "minBytes": 500,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/flake-evidence.md",
+            "pattern": "^##\\s+Symptom[\\s\\S]*^##\\s+Runs observed[\\s\\S]*^##\\s+Failure signature[\\s\\S]*^##\\s+Suspected class",
+            "flags": "im",
+            "label": "flake-evidence sections in order"
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "{{task.dir}}/flake-evidence.md",
+            "minCitations": 2,
+            "artifact": true
+          }
+        ]
+      },
+      {
+        "path": "{{task.dir}}/diagnosis.md",
+        "kind": "markdown-notes",
+        "artifact": true,
+        "minBytes": 600,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/diagnosis.md",
+            "pattern": "^##\\s+Mechanism[\\s\\S]*^##\\s+Experiment[\\s\\S]*^##\\s+Defect site[\\s\\S]*^##\\s+Siblings checked",
+            "flags": "im",
+            "label": "diagnosis sections in order"
+          },
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/diagnosis.md",
+            "pattern": "tests/flags\\.test\\.mjs|src/flags\\.js",
+            "label": "names the real site of the shared-state hazard"
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "{{task.dir}}/diagnosis.md",
+            "minCitations": 2,
+            "artifact": true
+          }
+        ]
+      },
+      {
+        "path": "{{task.dir}}/fix-notes.md",
+        "kind": "markdown-notes",
+        "artifact": true,
+        "minBytes": 700,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/fix-notes.md",
+            "pattern": "^##\\s+Problem[\\s\\S]*^##\\s+Change[\\s\\S]*^##\\s+Files touched[\\s\\S]*^##\\s+Regression coverage[\\s\\S]*^##\\s+Risk[\\s\\S]*^##\\s+How to verify",
+            "flags": "im",
+            "label": "fix-notes sections in order"
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "{{task.dir}}/fix-notes.md",
+            "minCitations": 2,
+            "artifact": true
+          }
+        ]
+      },
+      {
+        "path": "{{task.dir}}/validation.md",
+        "kind": "markdown-notes",
+        "artifact": true,
+        "minBytes": 300,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/validation.md",
+            "pattern": "^##\\s+Runs[\\s\\S]*^##\\s+Result",
+            "flags": "im",
+            "label": "validation sections in order"
+          }
+        ]
+      },
+      {
+        "path": "{{task.dir}}/review.md",
+        "kind": "markdown-report",
+        "artifact": true,
+        "minBytes": 400,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/review.md",
+            "pattern": "Verdict:\\s*(?:PASS|REVISE)",
+            "flags": "i",
+            "label": "explicit reviewer verdict"
+          }
+        ]
+      }
+    ],
+    "checks": [
+      {
+        "kind": "notContains",
+        "file": "tests/flags.test.mjs",
+        "pattern": "\\.skip\\s*\\(|\\.todo\\s*\\(|\\.only\\s*\\(|\\{\\s*skip\\s*:|\\{\\s*todo\\s*:",
+        "label": "the flaky test is not skipped, stubbed, or narrowed"
+      },
+      {
+        "kind": "sourceParses",
+        "file": "tests/flags.test.mjs"
+      },
+      {
+        "kind": "nodeRuns",
+        "file": "tests/flags.test.mjs",
+        "timeoutMs": 30000
+      },
+      {
+        "kind": "nodeScriptPasses",
+        "script": "tests/verify-deflake.mjs",
+        "timeoutMs": 60000,
+        "requiredOutput": [
+          {
+            "pattern": "DEFLAKE_ORACLE flags semantics ok",
+            "label": "the flags module still behaves"
+          },
+          {
+            "pattern": "DEFLAKE_ORACLE stable 5x",
+            "label": "five consecutive green suite runs"
+          },
+          {
+            "pattern": "DEFLAKE_ORACLE assertion intact",
+            "label": "the pristine-session assertion survives unweakened"
+          },
+          {
+            "pattern": "DEFLAKE_ORACLE order robust",
+            "label": "a late-appended test still sees a pristine session"
+          },
+          {
+            "pattern": "DEFLAKE_ORACLE mutant caught",
+            "label": "restoring the original order makes the suite fail again"
+          }
+        ]
+      }
+    ],
+    "taskNotes": {
+      "minBytes": 120,
+      "checks": [
+        {
+          "kind": "contains",
+          "file": "task-notes.md",
+          "pattern": "\\bDONE\\b[\\s\\S]*(npm\\s+run\\s+test|node\\s+--test|consecutive)",
+          "flags": "i",
+          "label": "DONE note cites the stability proof"
+        }
+      ]
+    },
+    "taskGraph": {
+      "requireCraftbookTask": true,
+      "requireTerminalStep": true
+    },
+    "unchangedFixtures": [
+      "package.json",
+      "src/flags.js",
+      "source/flake-report.md"
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "{{task.dir}}/fix-notes.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "Mechanism-level diagnosis",
+        "description": "The nondeterminism is named as a concrete mechanism (shared module state leaking across tests) and pinned with a real experiment, not guessed from the symptom."
+      },
+      {
+        "name": "No masking",
+        "description": "The flake is removed with cleanup or isolation that covers every test; the flaky test keeps its exact assertion and gains no skip, retry, or widened timeout."
+      },
+      {
+        "name": "Receipt-backed stability",
+        "description": "Stability is proven with consecutive green runs recorded as real receipts, and anything unverified is stated plainly instead of claimed."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "statistical reproduction",
+    "mechanism removal over masking",
+    "consecutive-green receipts"
+  ]
+}

package/data/craftbook-templates/fo/folder-to-audio/versions/1.0.3/craftbook.json

@@ -0,0 +1,215 @@
+{
+  "id": "folder-to-audio",
+  "name": "Text-to-Speech a Folder",
+  "description": "Turn a folder of text files (articles, chapters, notes, markdown) into a set of spoken audio files using text-to-speech, so a reading list becomes a listening list. Scopes the voice, file mapping, and chunking rules FIRST so long inputs do not get truncated or mispronounced, then synthesizes one audio artifact per source file with a consistent voice, then verifies every file was rendered and is audible. The scope-before-synthesize ordering is the point: deciding chunk size, file naming, and a manifest up front is what keeps a batch TTS run complete and re-runnable instead of half-finished.\n\nA gallery craftbook generated from an archetype spec. It runs\n`phase → (per-phase gate) → … → evaluate → (loop) → finish`. Each build\nphase that produces a checkable artifact is followed by a **runtime\ngate-checkpoint** — the runtime verifies the artifact and routes with no\nmodel turn, looping back to redo the phase on a miss. The final `evaluate`\nstep holds a static deliverable gate plus a reviewer QA pass. What it adds\nover the generic `build-loop`: a specialist role per phase, a\ndomain-correct ordering, and a concrete per-phase quality bar.\n\nDeliverables marked \"artifact\" land in the project's artifacts drawer (`write_artifact` / `read_artifact`), not the shipped workspace — review output is not product source.\n\nPhases:\n\n1. Scope the synthesis run (planner) — voice, file mapping, chunking, output manifest → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Synthesize the audio (developer) — render one audio file per source file with the locked voice → gated on `audio` (image-set)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "scope",
+  "triggers": [
+    "convert text to audio",
+    "text to speech a folder",
+    "narrate my articles",
+    "make audio from text files",
+    "tts batch"
+  ],
+  "paramSchema": {
+    "type": "object",
+    "properties": {
+      "workPath": {
+        "type": "string",
+        "title": "Working folder",
+        "description": "Per-task working folder in the artifacts drawer. Defaults to this task's own folder so runs never collide; override with a stable name when you deliberately want runs to share files.",
+        "default": "{{task.dir}}"
+      }
+    }
+  },
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the synthesis run",
+      "description": "voice, file mapping, chunking, output manifest",
+      "prompt": "Plan the batch BEFORE synthesizing anything. Step 1: list the source folder and record every text file (path, byte size). Step 2: pick ONE voice/engine (record the exact voice id and language) so all outputs sound consistent. Step 3: decide the input->output mapping (e.g. `article-01.md` -> `audio/article-01.mp3`) and the audio format/sample rate. Step 4: decide a chunking rule for long files (split on paragraph/sentence boundaries under the engine's character limit, then concatenate) so nothing is truncated. Step 5: decide pronunciation/normalization rules (expand numbers, abbreviations, URLs spoken-or-skipped). Write an acceptance-criteria checklist (one audio file per source file, all use the same voice, none is silent/zero-length, a manifest maps source->output->duration, long files are fully spoken not truncated) and the decisions to `{{workPath}}/scope.md` via write_task_note AND to the produces path.\n\nThe deliverable `{{workPath}}/scope.md` lands in the project's artifacts drawer — write it with `write_artifact` and read it back with `read_artifact`; the shipped workspace stays untouched.",
+      "suggestedRole": "planner",
+      "advanceWhen": {
+        "file": "{{workPath}}/scope.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/scope.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/scope.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "scope",
+        "maxAttempts": 3
+      },
+      "next": "synthesize",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "artifact",
+        "additionalOutputMedia": [
+          "task-note"
+        ]
+      }
+    },
+    {
+      "id": "synthesize",
+      "name": "Synthesize the audio",
+      "description": "render one audio file per source file with the locked voice",
+      "prompt": "Render the audio to satisfy the scope. Step 1: for each source file, read its text and apply the normalization rules from scope. Step 2: split into chunks under the engine's limit on sentence boundaries, synthesize each chunk with the locked voice, and concatenate into the single mapped output file under `audio/`. Step 3: write outputs incrementally (one finished file at a time) so progress is observable and a crash is resumable. Step 4: as you go, append a row to `audio/manifest.json` or a manifest table: source path, output path, voice id, approximate duration, byte size. Do not skip files; do not change voice mid-batch. On a loop-back, re-render only the files the reviewer flagged (missing, silent, truncated) and leave passing files untouched. write_task_note the audio directory path and a coverage count (rendered vs source files).",
+      "suggestedRole": "developer",
+      "advanceWhen": {
+        "file": "audio",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "fileCount",
+            "ext": [
+              "png",
+              "jpg",
+              "jpeg",
+              "svg",
+              "webp"
+            ],
+            "min": 3,
+            "dir": "audio"
+          }
+        ],
+        "onReject": "synthesize",
+        "maxAttempts": 4
+      },
+      "next": "evaluate",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "artifacts",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web"
+        ],
+        "outputMedium": "workspace",
+        "additionalOutputMedia": [
+          "task-note"
+        ]
+      }
+    },
+    {
+      "id": "evaluate",
+      "name": "Evaluate",
+      "description": "Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.",
+      "prompt": "Verify the batch is complete and audible. Step 1: count audio files in `audio/` vs source text files — every source must have an output (PASS/FAIL coverage). Step 2: confirm none is zero-length/silent (check byte size > a sane floor; sample-listen if an audio capability exists). Step 3: confirm the manifest maps each source->output with a plausible duration. Step 4: spot-check that the LONGEST source produced audio whose duration is consistent with its length (not truncated). Step 5: confirm all files share the locked voice/format. Write PASS/FAIL per criterion.\n\nThen route — this is the whole point of the loop:\n\n- **Every criterion PASSES →** call `advance_task_step({ ref, stepId: \"evaluate\", next: \"finish\" })`.\n- **Any criterion FAILS →** write the specific gaps to notes, then call `advance_task_step({ ref, stepId: \"evaluate\", next: \"synthesize\" })` to loop back. The builder fixes exactly those gaps.\n\nNever route to `finish` while any criterion is unmet. The build phase's completion gate already blocked a grossly-incomplete deliverable; your job is the judgment an automated check cannot make (does it actually work, read well, look right). After ~3 unproductive loops, stop and report DONE_WITH_CONCERNS so the user can step in.",
+      "suggestedRole": "reviewer",
+      "consumes": [
+        {
+          "file": "audio"
+        }
+      ],
+      "next": "synthesize",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "artifacts",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "task-note"
+      }
+    },
+    {
+      "id": "finish",
+      "name": "Finish",
+      "description": "All acceptance criteria met. Stamp a short summary and report DONE.",
+      "prompt": "Every acceptance criterion passed. Write a one-paragraph DONE summary to task notes via `write_task_note`: what was built, the deliverable path(s), and a one-line confirmation that each criterion is met. Then report DONE.",
+      "suggestedRole": "developer",
+      "terminal": true,
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "artifacts",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "task-note"
+      }
+    }
+  ],
+  "version": "1.0.3",
+  "releasedAt": "2026-09-01T20:00:00Z",
+  "minGezelVersion": "1.26233"
+}

package/data/craftbook-templates/fo/folder-to-audio/versions/1.0.3/test.json

@@ -0,0 +1,103 @@
+{
+  "schemaVersion": 1,
+  "title": "Text-to-Speech a Folder smoke eval",
+  "objective": "Self-contained smoke eval for the Text-to-Speech a Folder craftbook using the media generic harness.",
+  "tags": [
+    "media"
+  ],
+  "prompt": "Our media files are all listd in source/media-index.csv. Can you put together the production plan for them as media-plan.json?",
+  "setup": {
+    "projectName": "Text-to-Speech a Folder Eval",
+    "about": "Self-contained eval project for folder-to-audio. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/media-plan.json.",
+    "missionObjectives": "Use the Text-to-Speech a Folder craftbook/template, read the seeded local fixtures, and write media-plan.json without network calls, real credentials, or live services.",
+    "files": [
+      {
+        "path": "source/brief.md",
+        "content": "# Text-to-Speech a Folder Eval Brief\n\nClient: Boreal Desk, a home-office accessories company.\nAudience: operations leads who need an artifact they can use this week.\n\nFixed source facts for grounding:\n- The returns desk pilot covered 18 SKUs.\n- Median first response improved from 18 hours to 6 hours.\n- Preventable refund leakage fell from 14.2% to 8.9%.\n- The top unresolved complaint is status silence after photo submission.\n- Required next actions are automated status emails, barcode-exception training, and a weekly Finance exception export.\n\nUse these facts when the task asks for prose, analysis, copy, UI content, or test data. Do not use live web services, real credentials, or current outside data.\n\nCraftbook under test: folder-to-audio - Text-to-Speech a Folder.\n"
+      },
+      {
+        "path": "source/media-index.csv",
+        "content": "filename,type,duration_or_size,notes\npilot-interview.wav,audio,00:02:15,mentions response time and status silence\nreturns-desk.png,image,1600x900,shows intake form and photo evidence step\nlaunch-clip.mp4,video,00:01:05,shows barcode exception handling\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Producer"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "media-plan.json is a valid JSON media-production manifest grounded in seeded fixture metadata.",
+    "deliverables": [
+      {
+        "path": "media-plan.json",
+        "kind": "json",
+        "minBytes": 700,
+        "checks": [
+          {
+            "kind": "sniff",
+            "file": "media-plan.json",
+            "sniff": "json-valid"
+          },
+          {
+            "kind": "contains",
+            "file": "media-plan.json",
+            "pattern": "pilot-interview\\.wav|returns-desk\\.png|launch-clip\\.mp4",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "media-plan.json",
+            "pattern": "outputs|assets|steps|cues|verification",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "media-plan.json",
+            "pattern": "accessibility|alt|caption|transcript|metadata",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "media-plan.json",
+      "kind": "json"
+    },
+    "axes": [
+      {
+        "name": "completeness",
+        "description": "The manifest covers sources, outputs, and processing steps end to end."
+      },
+      {
+        "name": "grounding",
+        "description": "Only seeded media fixtures are referenced, with faithful metadata."
+      },
+      {
+        "name": "accessibility",
+        "description": "Alt text, captions, or transcripts are planned wherever the class requires."
+      },
+      {
+        "name": "precision",
+        "description": "Timestamps, crops, and cues are specific enough to execute."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "media fixture handling",
+    "structured manifest",
+    "verification metadata"
+  ],
+  "extensions": {
+    "legacySimulators": [
+      {
+        "id": "folder-to-audio-media-fixtures",
+        "kind": "data-source",
+        "status": "implemented",
+        "description": "Seeded local media index replacing real audio, video, image generation, and external media metadata services."
+      }
+    ]
+  }
+}