@bendyline/gilde 0.1.37 → 0.1.39

package/data/craftbook-templates/ar/artifact-integrity-review/versions/1.0.2/craftbook.json

@@ -0,0 +1,170 @@
+{
+  "id": "artifact-integrity-review",
+  "name": "Artifact Integrity & Ship Review",
+  "description": "Perform an isolated ship-readiness review of the actual built artifacts of a project—sites, npm packages, archives, CLIs, binaries, installers, checksums, signatures, and release bundles—rather than trusting source or CI status. Inventories expected versus present assets, stages immutable copies, inspects archive contents and metadata, runs safe sandboxed smoke checks where possible, and leaves per-artifact GO/NO-GO verdicts plus a release-wide decision. Can review locally staged assets or public-release evidence supplied to the project. Use before shipping a release or as a recurring artifact-integrity night-shift audit.\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. Inventory and stage the release (reviewer) — lock the expected matrix and immutable evidence set before inspecting it → gated on artifact `{{workPath}}/artifacts/inventory.md` (markdown-notes)\n2. Inspect each artifact (developer) — verify identity, contents, installability, safety, and cross-artifact consistency → gated on artifact `{{workPath}}/artifacts/checks.md` (markdown-notes)\n3. Write the artifact ship report (reviewer) — per-artifact verdicts and a release-wide ship decision → gated on artifact `{{workPath}}/artifact-integrity-review.md` (markdown-report)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "inventory",
+  "triggers": [
+    "artifact integrity review",
+    "review the built artifacts",
+    "test the release assets",
+    "are these installers ready to ship",
+    "audit a public github release",
+    "nightly artifact review"
+  ],
+  "toolsets": [
+    {
+      "toolsetId": "builtin.archives",
+      "autoAllow": true,
+      "reason": "inspect zip, tar, tgz, and package contents without trusting filenames"
+    },
+    {
+      "toolsetId": "builtin.code-execution",
+      "autoAllow": true,
+      "reason": "perform bounded, sandboxed metadata and smoke checks on staged artifacts"
+    },
+    {
+      "toolsetId": "builtin.web",
+      "autoAllow": true,
+      "reason": "read public release metadata and verify supplied public asset URLs when allowed"
+    }
+  ],
+  "steps": [
+    {
+      "id": "inventory",
+      "name": "Inventory and stage the release",
+      "description": "lock the expected matrix and immutable evidence set before inspecting it",
+      "prompt": "Define the exact release under review (version, commit/tag, target platforms, artifact types). Derive the expected asset matrix from release manifests, package metadata, CI/release configuration, checksums, and supplied release notes. Inventory every locally staged artifact with path, byte size, type/magic when available, and a cryptographic hash. If public release metadata or URLs were supplied and Web Access is allowed, verify the public listing and headers; never claim a remote binary was downloaded or inspected unless it is actually present in the workspace. Preserve the originals and work only on isolated extracted/staged copies. Record missing, duplicate, unexpected, truncated, or zero-byte assets immediately. Write the expected-vs-present matrix, provenance, hashes, acquisition limitations, and isolation plan to `{{workPath}}/artifacts/inventory.md`.\n\nThe deliverable `{{workPath}}/artifacts/inventory.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": "reviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/artifacts/inventory.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/artifacts/inventory.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/artifacts/inventory.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "inventory",
+        "maxAttempts": 3
+      },
+      "next": "inspect"
+    },
+    {
+      "id": "inspect",
+      "name": "Inspect each artifact",
+      "description": "verify identity, contents, installability, safety, and cross-artifact consistency",
+      "prompt": "Inspect every artifact independently. Use `list_archive` before extraction, then extract into isolated paths. Verify filename/type/magic, version and commit identity, checksums/signatures when provided, expected entrypoints, executable bits, package manifests, dependency/lock and license/SBOM notices, source-map/debug-symbol policy, update/uninstall metadata, and absence of secrets, private keys, `.env`, build caches, absolute developer paths, or unexpected source. Check sites for missing referenced assets. For npm/CLI/packages, perform only safe sandboxed pack/install/help/version/smoke checks; never run an untrusted native installer on the host. For platform artifacts that cannot run here, inspect statically and mark the platform smoke test NOT VERIFIED. Compare all artifacts for one version/commit and consistent bundled files. Record commands, exit codes, and evidence per artifact in `{{workPath}}/artifacts/checks.md`.\n\nThe deliverable `{{workPath}}/artifacts/checks.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": "developer",
+      "advanceWhen": {
+        "file": "{{workPath}}/artifacts/checks.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/artifacts/checks.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/artifacts/checks.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "inspect",
+        "maxAttempts": 3
+      },
+      "next": "report"
+    },
+    {
+      "id": "report",
+      "name": "Write the artifact ship report",
+      "description": "per-artifact verdicts and a release-wide ship decision",
+      "prompt": "Write `{{workPath}}/artifact-integrity-review.md` with: `## Release verdict`, `## Evidence provenance`, `## Artifact matrix`, `## Blocking findings`, `## Advisory findings`, `## Cross-artifact consistency`, `## Not verified`, and `## Reproduction commands`. Give every artifact its own GO / NO-GO / NOT-VERIFIED verdict. Every claim must point to a staged path, hash, manifest field, archive member, command/exit code, or supplied public-release record. Any missing expected artifact, secret/private key, checksum/signature mismatch, version/commit mismatch, broken entrypoint, failed smoke test, or missing referenced site asset is a blocker unless explicitly justified. A release cannot be GO when any required artifact is NO-GO or materially unverified. Keep source-code readiness outside scope except where needed to explain artifact provenance.\n\nThe deliverable `{{workPath}}/artifact-integrity-review.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": "reviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/artifact-integrity-review.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/artifact-integrity-review.md",
+            "bytes": 2000,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/artifact-integrity-review.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "report",
+        "maxAttempts": 4
+      },
+      "next": "evaluate"
+    },
+    {
+      "id": "evaluate",
+      "name": "Evaluate",
+      "description": "Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.",
+      "prompt": "Grade `{{workPath}}/artifact-integrity-review.md`: (1) expected and present assets are reconciled; (2) each artifact has provenance, size/type/hash, checks, and an individual verdict; (3) archive/package contents and metadata were inspected in isolation; (4) safe smoke tests include commands and exit codes while unrun platform tests are NOT VERIFIED; (5) versions/commits and bundled content agree across artifacts; (6) secrets, licenses/SBOM, source maps/debug files, signatures/checksums, entrypoints, and missing site assets were considered where applicable; (7) the release verdict follows blockers and uncertainty. PASS all or loop back with exact gaps.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/artifact-integrity-review.md` with `read_artifact`, not `read_file`.\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: \"report\" })` 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": "{{workPath}}/artifact-integrity-review.md",
+          "artifact": true
+        }
+      ],
+      "next": "report"
+    },
+    {
+      "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
+    }
+  ],
+  "version": "1.0.2",
+  "releasedAt": "2026-08-21T00:00:00Z",
+  "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}}"
+      }
+    }
+  },
+  "minGezelVersion": "1.26233"
+}

package/data/craftbook-templates/ar/artifact-integrity-review/versions/1.0.2/test.json

@@ -0,0 +1,139 @@
+{
+  "schemaVersion": 1,
+  "title": "Artifact Integrity & Ship Review isolated eval",
+  "objective": "Measure whether the review reconciles expected and staged release assets, catches cross-artifact identity and content blockers, and refuses an unsafe GO verdict.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "Use the Artifact Integrity & Ship Review craftbook on the staged release/ tree. Treat release/expected.json as the release contract, inspect only local evidence, and write tasks/eval/artifact-integrity-review.md. Do not modify or execute the artifacts.",
+  "setup": {
+    "projectName": "Artifact Integrity Maintenance Eval",
+    "about": "An extracted release fixture with deliberate missing, mismatched, secret-bearing, and broken-site assets.",
+    "missionObjectives": "Give each expected artifact an honest verdict and issue a release-wide NO-GO when blockers remain.",
+    "files": [
+      {
+        "path": "release/expected.json",
+        "content": "{\n  \"version\": \"2.4.0\",\n  \"commit\": \"abc123\",\n  \"required\": [\"site\", \"cli-package\", \"windows-installer.exe\"],\n  \"siteEntry\": \"site/index.html\",\n  \"cliVersion\": \"2.4.0\"\n}\n"
+      },
+      {
+        "path": "release/site/index.html",
+        "content": "<!doctype html><html><body><main>Returns Desk 2.4.0</main><script src=\"app.abc.js\"></script></body></html>\n"
+      },
+      {
+        "path": "release/site/app.def.js",
+        "content": "console.log('build abc123');\n"
+      },
+      {
+        "path": "release/cli-package/package.json",
+        "content": "{\"name\":\"returns-cli\",\"version\":\"2.3.9\",\"bin\":{\"returns\":\"bin/returns.js\"},\"files\":[\"bin\",\".env\"]}\n"
+      },
+      {
+        "path": "release/cli-package/bin/returns.js",
+        "content": "#!/usr/bin/env node\nconsole.log('returns-cli 2.3.9');\n"
+      },
+      {
+        "path": "release/cli-package/.env",
+        "content": "RETURNS_API_TOKEN=fixture-secret-must-not-ship\n"
+      },
+      {
+        "path": "release/checksums.txt",
+        "content": "0000000000000000000000000000000000000000000000000000000000000000  site/index.html\n"
+      }
+    ],
+    "worker": {
+      "name": "Rae",
+      "role": "Reviewer"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "tasks/eval/artifact-integrity-review.md issues NO-GO, reconciles all expected assets, and cites each seeded blocker and uncertainty.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/artifact-integrity-review.md",
+        "kind": "markdown-report",
+        "minBytes": 1700,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "NO-GO|NO GO|block",
+            "flags": "i",
+            "label": "release verdict"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "windows-installer\\.exe[\\s\\S]{0,160}(missing|not present|NO-GO)",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "app\\.abc\\.js[\\s\\S]{0,160}(missing|broken)|app\\.def\\.js",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "2\\.3\\.9[\\s\\S]{0,160}2\\.4\\.0|version mismatch",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "\\.env|RETURNS_API_TOKEN|secret",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "checksum|000000",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "GO|NO-GO|NOT-VERIFIED",
+            "flags": "i",
+            "label": "per-artifact verdict vocabulary"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/artifact-integrity-review.md",
+            "pattern": "Not verified|NOT-VERIFIED|platform",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/artifact-integrity-review.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "reconciliation",
+        "description": "Expected and staged assets are exhaustively matched with provenance."
+      },
+      {
+        "name": "isolation and evidence",
+        "description": "Checks rely on staged paths, metadata, hashes, and bounded local inspection."
+      },
+      {
+        "name": "verdict integrity",
+        "description": "Per-artifact and release-wide verdicts follow blockers and uncertainty."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "expected-vs-present inventory",
+    "cross-artifact identity",
+    "honest ship verdict"
+  ]
+}

package/data/craftbook-templates/au/audio-ad-spot/versions/1.0.2/craftbook.json

@@ -0,0 +1,141 @@
+{
+  "id": "audio-ad-spot",
+  "name": "Audio Ad Spot",
+  "description": "Produce a short broadcast/podcast audio ad spot (15s/30s/60s) — a tightly-timed script with a hook, value, and a single call-to-action, synthesized into a finished, runtime-accurate voiced spot with optional music bed cue. Scopes the offer, exact runtime, and mandatory disclaimers FIRST, then synthesizes a voiced spot that hits the time target. Locking the runtime and CTA before recording is what makes a spot land in its slot and drive one action instead of overrunning or rambling.\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. Ad scope (planner) — offer, exact runtime, single CTA, disclaimers, voice → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Write the spot script (copywriter) — timed ad script with hook, CTA, disclaimer, cues → gated on artifact `{{workPath}}/ad-script.md` (markdown-notes)\n3. Produce the spot (developer) — synthesize the voiced spot to the exact runtime → gated on `ad-spot.mp3` (audio-file)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "scope",
+  "triggers": [
+    "make a radio ad",
+    "audio ad spot",
+    "30 second commercial",
+    "podcast ad read",
+    "produce an audio ad"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Ad scope",
+      "description": "offer, exact runtime, single CTA, disclaimers, voice",
+      "prompt": "Lock the spot BEFORE writing or voicing. Step 1: capture the product/offer, the ONE target action (the single CTA), and the audience. Step 2: pick the EXACT runtime slot (15/30/60s) and derive the word budget (~2.5 words/second for ad reads — record it). Step 3: capture any mandatory disclaimer/legal line and how the brand/CTA must be stated. Step 4: pick the voice and tone, and whether a music bed cue is wanted. Step 5: define the arc: hook (first ~3s) -> value -> CTA -> brand/disclaimer. Write an acceptance-criteria checklist (finished audio within ±1s of the slot, opens with a hook in the first 3s, states exactly one CTA, includes the disclaimer, brand named clearly, not silent) and decisions to `{{workPath}}/scope.md` via write_task_note AND 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": "script"
+    },
+    {
+      "id": "script",
+      "name": "Write the spot script",
+      "description": "timed ad script with hook, CTA, disclaimer, cues",
+      "prompt": "Write the spot to the scope. Step 1: write to the word budget so the read lands in the slot — open with the hook in the first ~3 seconds. Step 2: keep one clear value message and exactly ONE CTA, stated plainly. Step 3: include the brand name and the mandatory disclaimer (a quick read at the end is fine, but it counts against the time). Step 4: add VO cues (`[pause]`, emphasis, pace) and a music-bed cue if scoped, and write numbers/phone/URLs as they should be SPOKEN. Step 5: write the script to `{{workPath}}/ad-script.md` with a running word count and estimated runtime via write_task_note AND the produces path. On a loop-back, trim/expand only to fix the time or a missing element.\n\nThe deliverable `{{workPath}}/ad-script.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": "copywriter",
+      "advanceWhen": {
+        "file": "{{workPath}}/ad-script.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/ad-script.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/ad-script.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "script",
+        "maxAttempts": 3
+      },
+      "next": "produce"
+    },
+    {
+      "id": "produce",
+      "name": "Produce the spot",
+      "description": "synthesize the voiced spot to the exact runtime",
+      "prompt": "Render the finished spot. Step 1: synthesize the approved script with the locked voice/tone via the text-to-speech capability. Step 2: measure the rendered duration; if it is outside ±1s of the slot, adjust pacing or trim the script (loop the copy phase if it cannot fit) until it lands. Step 3: if a music bed was scoped, mix a low ducked bed under the VO and apply gentle fades. Step 4: normalize loudness to a broadcast-sane level so it is not too quiet/hot. Step 5: write the finished audio to `ad-spot.mp3` and note its exact measured duration. write_task_note the path and duration. On a loop-back, fix only the flagged issue (over/under time, missing CTA/disclaimer, silence).",
+      "suggestedRole": "developer",
+      "advanceWhen": {
+        "file": "ad-spot.mp3",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "ad-spot.mp3",
+            "bytes": 1000
+          }
+        ],
+        "onReject": "produce",
+        "maxAttempts": 4
+      },
+      "next": "evaluate"
+    },
+    {
+      "id": "evaluate",
+      "name": "Evaluate",
+      "description": "Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.",
+      "prompt": "Time it and listen. Step 1: confirm `ad-spot.mp3` exists, is non-silent, and its duration is within ±1s of the target slot (PASS/FAIL). Step 2: sample-listen (if audio capability exists) or check the script — does it open with a hook in the first ~3s? Step 3: confirm exactly ONE CTA is stated and the brand is named clearly. Step 4: confirm the mandatory disclaimer is present. Step 5: confirm loudness is reasonable (and the music bed, if any, is ducked under the VO). 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: \"produce\" })` 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": "ad-spot.mp3"
+        }
+      ],
+      "next": "produce"
+    },
+    {
+      "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
+    }
+  ],
+  "version": "1.0.2",
+  "releasedAt": "2026-08-21T00:00:00Z",
+  "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}}"
+      }
+    }
+  },
+  "minGezelVersion": "1.26233"
+}

package/data/craftbook-templates/au/audio-ad-spot/versions/1.0.2/test.json

@@ -0,0 +1,106 @@
+{
+  "schemaVersion": 1,
+  "title": "Audio Ad Spot smoke eval",
+  "objective": "Self-contained smoke eval for the Audio Ad Spot 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": "Audio Ad Spot Eval",
+    "about": "Self-contained eval project for audio-ad-spot. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/media-plan.json.",
+    "missionObjectives": "Use the Audio Ad Spot 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": "# Audio Ad Spot 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: audio-ad-spot - Audio Ad Spot.\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"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "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": "audio-ad-spot-media-fixtures",
+        "kind": "data-source",
+        "status": "implemented",
+        "description": "Seeded local media index replacing real audio, video, image generation, and external media metadata services."
+      }
+    ]
+  }
+}