@bendyline/gilde 0.1.38 → 0.1.40

package/data/craftbook-templates/mo/morning-report/versions/1.0.2/test.json

@@ -0,0 +1,253 @@
+{
+  "schemaVersion": 1,
+  "title": "Morning House Report smoke eval",
+  "objective": "Task-specific eval for the Morning House Report craftbook: the whole fake house is polled through the wired device tools, the garage door left open overnight and the low sensor battery surface first with suggested actions, and the nominal rest is compressed to a glanceable line or two.",
+  "tags": [
+    "external"
+  ],
+  "prompt": "morning - whats the house looking like today? give me the unusal stuff first",
+  "setup": {
+    "projectName": "Smart Home — Morning Report Eval",
+    "about": "Self-contained eval project for morning-report. The fake house's device tools are wired into the project's tool roster — the device inventory and the live state snapshot come from them, and they are the only source of truth for what the house is doing this morning. The polling evidence goes to tasks/eval/states.md and the morning report to tasks/eval/report.md.",
+    "missionObjectives": "Use the Morning House Report craftbook/template, read every zone's state through the wired device tools, surface anything unusual first with one suggested action each, compress the nominal rest, and write tasks/eval/report.md. No real smart-home services, hubs, or credentials — only the wired fake house.",
+    "files": [],
+    "worker": {
+      "name": "Roos",
+      "role": "Huisautomaat"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [
+    {
+      "kind": "mcp",
+      "id": "home",
+      "description": "Fake smart-home MCP for this eval. Its tools surface directly on the tool roster: list the house's device inventory and read the current state snapshot of every device, battery levels included.",
+      "tools": [
+        {
+          "name": "list_devices",
+          "description": "List every device in the house with its id, name, zone, and type",
+          "resultTemplate": {
+            "devices": [
+              {
+                "id": "dev-01",
+                "name": "porch light",
+                "zone": "outside",
+                "type": "light"
+              },
+              {
+                "id": "dev-02",
+                "name": "front hall presence sensor",
+                "zone": "hall",
+                "type": "presence"
+              },
+              {
+                "id": "dev-03",
+                "name": "living room thermostat",
+                "zone": "living room",
+                "type": "thermostat"
+              },
+              {
+                "id": "dev-04",
+                "name": "garage door",
+                "zone": "garage",
+                "type": "cover"
+              },
+              {
+                "id": "dev-05",
+                "name": "kitchen ceiling light",
+                "zone": "kitchen",
+                "type": "light"
+              },
+              {
+                "id": "dev-06",
+                "name": "bedroom lamp",
+                "zone": "bedroom",
+                "type": "light"
+              },
+              {
+                "id": "dev-07",
+                "name": "front door lock",
+                "zone": "hall",
+                "type": "lock"
+              },
+              {
+                "id": "dev-08",
+                "name": "backyard motion sensor",
+                "zone": "outside",
+                "type": "motion"
+              }
+            ]
+          }
+        },
+        {
+          "name": "get_state",
+          "description": "Read the current state snapshot of every device in the house, with battery levels and last-change times where the device reports them",
+          "resultTemplate": {
+            "asOf": "2026-07-19T06:40:00Z",
+            "states": [
+              {
+                "id": "dev-01",
+                "name": "porch light",
+                "state": "off",
+                "reachable": true
+              },
+              {
+                "id": "dev-02",
+                "name": "front hall presence sensor",
+                "state": "home",
+                "reachable": true
+              },
+              {
+                "id": "dev-03",
+                "name": "living room thermostat",
+                "state": "heat 19.5C",
+                "reachable": true
+              },
+              {
+                "id": "dev-04",
+                "name": "garage door",
+                "state": "open",
+                "since": "2026-07-18T23:41:00Z",
+                "reachable": true
+              },
+              {
+                "id": "dev-05",
+                "name": "kitchen ceiling light",
+                "state": "off",
+                "reachable": true
+              },
+              {
+                "id": "dev-06",
+                "name": "bedroom lamp",
+                "state": "off",
+                "reachable": true
+              },
+              {
+                "id": "dev-07",
+                "name": "front door lock",
+                "state": "locked",
+                "reachable": true
+              },
+              {
+                "id": "dev-08",
+                "name": "backyard motion sensor",
+                "state": "clear",
+                "battery": 9,
+                "reachable": true
+              }
+            ]
+          }
+        }
+      ]
+    }
+  ],
+  "success": {
+    "summary": "tasks/eval/report.md leads with the garage door open since 23:41 and the backyard sensor at 9 percent battery — each with a suggested action — before any all-clear, the evidence in tasks/eval/states.md carries the real device ids from the tools (inventory and state snapshot actually read — request-log assertions), and the nominal rest is compressed rather than dumped.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/states.md",
+        "kind": "markdown-notes",
+        "minBytes": 200,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/states.md",
+            "pattern": "dev-\\d{2}",
+            "label": "the state evidence carries real device ids from the tools"
+          }
+        ]
+      },
+      {
+        "path": "tasks/eval/report.md",
+        "kind": "markdown-report",
+        "minBytes": 400,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "attention|unusual|anomal",
+            "flags": "i",
+            "label": "an anomalies-first section exists"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "garage",
+            "flags": "i",
+            "label": "the garage door left open overnight is surfaced"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "batter",
+            "flags": "i",
+            "label": "the low sensor battery is surfaced"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "close|shut|charge|replace|check",
+            "flags": "i",
+            "label": "a suggested action is offered"
+          },
+          {
+            "kind": "judge",
+            "file": "tasks/eval/report.md",
+            "rubric": "The two anomalies (the garage door open overnight and the backyard sensor's low battery) appear before anything normal and each carries one concrete suggested action, every device mentioned matches the recorded state evidence, the nominal rest of the house is compressed to a line or two rather than a device-by-device dump, and the whole report reads comfortably in under a minute.",
+            "sourceFiles": [
+              "tasks/eval/states.md"
+            ],
+            "label": "anomalies-first and glanceable"
+          }
+        ]
+      }
+    ],
+    "mocks": [
+      {
+        "service": "home",
+        "minRequests": 2,
+        "requiredTools": [
+          "list_devices",
+          "get_state"
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/report.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "anomaly-detection",
+        "description": "The open garage door and the low battery are both caught from the state snapshot; nothing unusual is missed or invented."
+      },
+      {
+        "name": "prioritization",
+        "description": "Anomalies lead, most urgent first, each with one concrete suggested action; the all-clear never buries them."
+      },
+      {
+        "name": "brevity",
+        "description": "The nominal rest is compressed by zone to a line or two — glanceable before coffee, no inventory dump."
+      },
+      {
+        "name": "warmth",
+        "description": "It reads like the house saying good morning — plain, friendly words, no telemetry jargon."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "anomalies surfaced first",
+    "grounded device states",
+    "glanceable brevity"
+  ]
+}

package/data/craftbook-templates/mu/music-metadata-tag/manifest.json

@@ -3,6 +3,7 @@
   "kind": "craftbook-template",
   "id": "music-metadata-tag",
   "role": "general",
+  "category": "media",
   "name": "Music Library Metadata Tagging",
   "description": "Tag and organize a folder of audio files with consistent, correct metadata — title, artist, album, track number, year, genre — so the library is clean, sortable, and player-friendly. Locks the tag schema and naming/foldering convention FIRST, applies tags and a consistent file/folder layout across the whole library, then verifies every file is fully and consistently tagged. Defining the schema and the source of truth before touching files is what prevents a half-tagged, inconsistently-named mess.",
   "tags": [

package/data/craftbook-templates/mu/music-metadata-tag/versions/1.0.2/craftbook.json

@@ -0,0 +1,113 @@
+{
+  "id": "music-metadata-tag",
+  "name": "Music Library Metadata Tagging",
+  "description": "Tag and organize a folder of audio files with consistent, correct metadata — title, artist, album, track number, year, genre — so the library is clean, sortable, and player-friendly. Locks the tag schema and naming/foldering convention FIRST, applies tags and a consistent file/folder layout across the whole library, then verifies every file is fully and consistently tagged. Defining the schema and the source of truth before touching files is what prevents a half-tagged, inconsistently-named mess.\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. Tag schema & convention (planner) — required fields, vocab, file/folder naming rules → gated on artifact `{{workPath}}/schema.md` (markdown-notes)\n2. Apply tags & layout (developer) — write metadata + rename/foldering per the schema → gated on `tag-report.csv` (data-file)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "schema",
+  "triggers": [
+    "tag my music",
+    "fix audio metadata",
+    "organize a music library",
+    "id3 tags",
+    "clean up song files"
+  ],
+  "steps": [
+    {
+      "id": "schema",
+      "name": "Tag schema & convention",
+      "description": "required fields, vocab, file/folder naming rules",
+      "prompt": "Define the schema BEFORE tagging. Step 1: list the audio folder and record every file (path, format, current tags if readable). Step 2: define the required tag fields (title, artist, album, album-artist, track number, disc, year, genre) and which are mandatory. Step 3: define controlled vocabularies/formatting: genre list, title-case rules, how 'feat.' is written, track-number padding (01, 02…). Step 4: define the file/folder convention (e.g. `Artist/Album/01 - Title.ext`). Step 5: name the source of truth for correcting tags (existing tags, filename parsing, or a lookup). Write an acceptance-criteria checklist (every file has all mandatory fields, genres drawn from the vocab, track numbers padded and sequential within an album, files follow the naming convention, a report lists before/after per file) and decisions to `{{workPath}}/schema.md` via write_task_note AND the produces path.\n\nThe deliverable `{{workPath}}/schema.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}}/schema.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/schema.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/schema.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "schema",
+        "maxAttempts": 3
+      },
+      "next": "tag"
+    },
+    {
+      "id": "tag",
+      "name": "Apply tags & layout",
+      "description": "write metadata + rename/foldering per the schema",
+      "prompt": "Apply the schema across the library. Step 1: for each file, derive correct field values from the source of truth, normalizing to the controlled vocab and formatting rules. Step 2: write the metadata tags to the file. Step 3: rename/move the file to match the naming convention, padding track numbers and grouping into Artist/Album folders. Step 4: process incrementally and append a row to `tag-report.csv` per file: old path, new path, and the final field values — so the run is observable and re-runnable. Step 5: do not skip files; flag any you cannot confidently tag rather than guessing wildly. write_task_note the report path and a coverage count. On a loop-back, re-tag only the flagged files.",
+      "suggestedRole": "developer",
+      "advanceWhen": {
+        "file": "tag-report.csv",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "tag-report.csv",
+            "bytes": 120
+          },
+          {
+            "kind": "sniff",
+            "file": "tag-report.csv",
+            "sniff": "data-table"
+          }
+        ],
+        "onReject": "tag",
+        "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": "Audit the tagged library. Step 1: confirm `tag-report.csv` covers every source file (count match) — PASS/FAIL coverage. Step 2: confirm every file has all MANDATORY fields populated. Step 3: confirm genres come from the controlled vocab and track numbers are padded and sequential within each album. Step 4: confirm files follow the naming convention (Artist/Album/NN - Title). Step 5: spot-check 3 files by reading back their tags — do they match the report? 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: \"tag\" })` 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": "tag-report.csv"
+        }
+      ],
+      "next": "tag"
+    },
+    {
+      "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/mu/music-metadata-tag/versions/1.0.2/test.json

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

package/data/craftbook-templates/na/narrated-slideshow/manifest.json

@@ -3,6 +3,7 @@
   "kind": "craftbook-template",
   "id": "narrated-slideshow",
   "role": "general",
+  "category": "media",
   "name": "Animated Content Slideshow",
   "description": "Turn written content into a real MP4 video and animated GIF with DocBlocks. Locks the story and one message per scene, writes one H1 per scene in slideshow.md, then performs one multi-target conversion, previews the frames, and saves both rendered files in the artifacts drawer — no hand-built HTML player.",
   "tags": [