@bendyline/gilde 0.1.38 → 0.1.39

package/data/craftbook-templates/pr/practice-session/versions/1.0.2/craftbook.json

@@ -0,0 +1,127 @@
+{
+  "id": "practice-session",
+  "name": "Log the Practice Session",
+  "description": "Close out today's practice properly: log what was actually worked with honest minutes and tempo marks, and set up tomorrow's session so the next sit-down starts with intent instead of noodling. Locks the honest picture of the session FIRST, then logs it into the practice record, then writes tomorrow's targeted plan. Use for logging a practice session, ending today's practice, planning tomorrow's practice, and 'log my practice'.\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 session (coach) — honest picture of today: pieces, minutes, tempos, trouble spots → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Log the session (coach) — today's entry into the practice record, piece states updated\n3. Write the close-out (coach) — today in two lines, streak state, tomorrow's targeted plan → gated on artifact `{{workPath}}/report.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": "scope",
+  "triggers": [
+    "log my practice",
+    "log today's session",
+    "just finished practicing",
+    "set up tomorrow's practice",
+    "close out the practice session"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the session",
+      "description": "honest picture of today: pieces, minutes, tempos, trouble spots",
+      "prompt": "Get the honest picture of today's session before logging a minute of it. Step 1: read practice.json and repertoire.json from the workspace — the streak, the pieces in rotation, and where each one stands. Step 2: confirm with the owner what actually happened today: which pieces, honest minutes (the ones at the instrument, not the ones planned), tempo marks reached, and the spots that fought back. Step 3: write an acceptance-criteria checklist ('today's session is logged with date, minutes, and piece', 'trouble spots recorded by bar or passage, not vibes', 'piece states in the repertoire updated where today changed them', 'tomorrow's plan targets the trouble spots with concrete tempo marks'). Call write_task_note with the minutes + pieces + checklist and write {{workPath}}/scope.md.\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": "coach",
+      "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": "log"
+    },
+    {
+      "id": "log",
+      "name": "Log the session",
+      "description": "today's entry into the practice record, piece states updated",
+      "prompt": "Log today exactly as it went. Step 1: record the session through the project's practice capabilities — date, minutes, piece(s), tempo marks, and what felt solid versus mushy, in the owner's own words. Step 2: update the piece states in the repertoire where today moved them — forward or backward; honest beats flattering. Step 3: confirm the log landed by reading practice.json back — today's entry must be there with its date, minutes, and piece. Call write_task_note with the entry as logged.",
+      "suggestedRole": "coach",
+      "next": "report"
+    },
+    {
+      "id": "report",
+      "name": "Write the close-out",
+      "description": "today in two lines, streak state, tomorrow's targeted plan",
+      "prompt": "Close the day and open tomorrow. Step 1: write today in two honest lines — what got the minutes and how it went. Step 2: state the streak — how many days running, and what today did to it. Step 3: write tomorrow's plan: the specific passages to hit (by bar number), the metronome mark to start at and the mark to reach for, and one thing to leave alone because it is already working. Step 4: keep it encouraging and true — a coach's note, not a report card. Write {{workPath}}/report.md and call write_task_note with tomorrow's top target.\n\nThe deliverable `{{workPath}}/report.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": "coach",
+      "advanceWhen": {
+        "file": "{{workPath}}/report.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/report.md",
+            "bytes": 1000,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/report.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": "Open {{workPath}}/report.md and practice.json and verify every criterion from {{workPath}}/scope.md: today's entry is in the log with date, minutes, and piece matching what the owner reported; trouble spots are recorded by bar or passage; repertoire states moved where today moved them; and tomorrow's plan names specific passages with metronome marks. Write PASS/FAIL per criterion; on any failure, name the gap and loop back to report.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/report.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}}/report.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/pr/practice-session/versions/1.0.2/test.json

@@ -0,0 +1,117 @@
+{
+  "schemaVersion": 1,
+  "title": "Log the Practice Session — 40 minutes on the nocturne",
+  "objective": "Measure whether the practice-session craftbook logs today's session into the practice record with honest specifics and sets up a targeted plan for tomorrow.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "just did 40 min on the chopin nocturne, tempo work went ok but bar 32 is still mushy — log it and set up tomorows session",
+  "setup": {
+    "projectName": "Piano Practice",
+    "about": "Music practice project. The session log is workspace/practice.json (one entry per session); the pieces in rotation live in workspace/repertoire.json.",
+    "missionObjectives": "Keep an honest practice log and end every session with tomorrow already planned. Grounded in the seeded log and repertoire; no live services, no real credentials.",
+    "files": [
+      {
+        "path": "practice.json",
+        "content": "[\n  {\n    \"date\": \"2026-07-11\",\n    \"minutes\": 35,\n    \"piece\": \"Chopin Nocturne Op. 9 No. 2\",\n    \"tempo\": \"88 bpm hands together\",\n    \"tasks/eval\": \"bars 1-16 solid, ornaments in bar 12 uneven\"\n  },\n  {\n    \"date\": \"2026-07-12\",\n    \"minutes\": 25,\n    \"piece\": \"Bach Invention No. 8\",\n    \"tempo\": \"72 bpm\",\n    \"tasks/eval\": \"left hand alone, felt good\"\n  },\n  {\n    \"date\": \"2026-07-14\",\n    \"minutes\": 45,\n    \"piece\": \"Chopin Nocturne Op. 9 No. 2\",\n    \"tempo\": \"92 bpm\",\n    \"tasks/eval\": \"bar 32 run falls apart above 90\"\n  },\n  {\n    \"date\": \"2026-07-17\",\n    \"minutes\": 30,\n    \"piece\": \"scales and arpeggios\",\n    \"tempo\": \"quarter = 100\",\n    \"tasks/eval\": \"E-flat major still lumpy on the way down\"\n  },\n  {\n    \"date\": \"2026-07-18\",\n    \"minutes\": 40,\n    \"piece\": \"Chopin Nocturne Op. 9 No. 2\",\n    \"tempo\": \"94 bpm\",\n    \"tasks/eval\": \"bar 32 slightly better at 85, breaks at 94\"\n  }\n]\n"
+      },
+      {
+        "path": "repertoire.json",
+        "content": "{\n  \"pieces\": [\n    {\n      \"piece\": \"Chopin Nocturne Op. 9 No. 2\",\n      \"state\": \"tempo building\",\n      \"currentTempo\": \"94 bpm\",\n      \"targetTempo\": \"116 bpm\",\n      \"troubleSpots\": [\"bar 12 ornaments\", \"bar 32 run\"]\n    },\n    {\n      \"piece\": \"Bach Invention No. 8\",\n      \"state\": \"hands separate\",\n      \"currentTempo\": \"72 bpm\",\n      \"targetTempo\": \"96 bpm\",\n      \"troubleSpots\": []\n    },\n    {\n      \"piece\": \"scales and arpeggios\",\n      \"state\": \"maintenance\",\n      \"currentTempo\": \"quarter = 100\",\n      \"targetTempo\": \"quarter = 120\",\n      \"troubleSpots\": [\"E-flat major descending\"]\n    }\n  ]\n}\n"
+      }
+    ],
+    "worker": {
+      "name": "Ruben",
+      "role": "Coach"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "Today's session is appended to practice.json as a well-formed entry, and tasks/eval/report.md closes the day with the streak state and a concrete, targeted plan for tomorrow.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/report.md",
+        "kind": "markdown-report",
+        "minBytes": 800,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "tomorrow|next session",
+            "flags": "i",
+            "label": "tomorrow's plan present"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "bpm|bar\\s*\\d+",
+            "flags": "i",
+            "label": "concrete tempo or bar targets"
+          }
+        ]
+      }
+    ],
+    "checks": [
+      {
+        "kind": "recordSchema",
+        "file": "practice.json",
+        "fields": [
+          {
+            "name": "date",
+            "type": "iso-date",
+            "required": true
+          },
+          {
+            "name": "minutes",
+            "type": "number",
+            "required": true
+          },
+          {
+            "name": "piece",
+            "type": "nonempty",
+            "required": true
+          }
+        ],
+        "minRows": 6
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/report.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "honesty",
+        "description": "The log records what actually happened — 40 minutes, tempo where it broke, bar 32 still mushy — without inflation."
+      },
+      {
+        "name": "specificity",
+        "description": "Trouble spots by bar number, tempos as metronome marks, never vague 'went ok'."
+      },
+      {
+        "name": "encouragement",
+        "description": "Reads like a coach's note — warm, forward-looking — not a report card."
+      },
+      {
+        "name": "plan quality",
+        "description": "Tomorrow's plan targets the actual trouble spots with a starting mark and a reach mark, and leaves working things alone."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "honest logging",
+    "targeted planning",
+    "record integrity"
+  ]
+}

package/data/craftbook-templates/pr/pre-departure-countdown/versions/1.0.2/craftbook.json

@@ -0,0 +1,152 @@
+{
+  "id": "pre-departure-countdown",
+  "name": "Pre-Departure Countdown",
+  "description": "Audit a trip's readiness in the final stretch before departure: verify documents and bookings, close the packing gaps against the packing roster, sweep home-prep tasks, and produce a day-by-day countdown checklist to the departure date — so the last week runs on a list, not on 3am worry. Locks the audit areas and the days remaining FIRST, then audits each area against the trip record, then writes the countdown. Use for pre-trip checks, departure prep, 'are we ready for the trip', and travel countdown checklists.\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 countdown (planner) — days remaining, audit areas, readiness criteria → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Audit readiness (reviewer) — check each area, grow the packing roster, list open items → gated on artifact `{{workPath}}/audit.md` (markdown-notes)\n3. Write the countdown (copywriter) — day-by-day checklist to departure → gated on artifact `{{workPath}}/report.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": "scope",
+  "triggers": [
+    "are we ready for the trip",
+    "pre departure checklist",
+    "trip countdown",
+    "before we leave",
+    "departure prep"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the countdown",
+      "description": "days remaining, audit areas, readiness criteria",
+      "prompt": "Define the audit before checking anything. Step 1: read trip.json and packing.json from the workspace — the destination, the departure date and days remaining, the planned days, and the packing roster's packed/unpacked split. Step 2: fix the audit areas — documents (passports/IDs valid, visas if needed), bookings (transport, lodging, the first day's plan), packing (the roster, plus anything the itinerary implies that isn't on it), money and phones (cards that work there, plans that roam), and home prep (mail, plants, locks, who has a key). Step 3: agree the readiness criterion per area — what 'checked' concretely means. Step 4: write an acceptance-criteria checklist ('every area audited with its criterion', 'every itinerary-implied item is on the packing roster', 'every open item lands on a specific countdown day', 'the checklist ends at walk-out-the-door'). Call write_task_note with days-remaining + areas + checklist and write {{workPath}}/scope.md.\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": "audit"
+    },
+    {
+      "id": "audit",
+      "name": "Audit readiness",
+      "description": "check each area, grow the packing roster, list open items",
+      "prompt": "Audit each area against its criterion. Step 1: walk documents, bookings, money/phones, and home prep with the owner — record each as ready or open-with-what's-missing. Step 2: read the itinerary's planned days and derive what they imply for packing (boots for the walking day, adapters, medications for the length of the trip); add anything missing to the packing roster through the project's trip capabilities. Step 3: list the packing roster's remaining unpacked items. Step 4: for every open item, capture the concrete closing action and how long it takes (some need a week — expedited documents; some need an hour). Step 5: write the audit results to {{workPath}}/audit.md and call write_task_note with ready/open counts per area.\n\nThe deliverable `{{workPath}}/audit.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}}/audit.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/audit.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/audit.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "audit",
+        "maxAttempts": 3
+      },
+      "next": "countdown"
+    },
+    {
+      "id": "countdown",
+      "name": "Write the countdown",
+      "description": "day-by-day checklist to departure",
+      "prompt": "Write the day-by-day countdown from the audit. Step 1: place every open item and every unpacked-item batch onto a specific day between now and departure, longest-lead-time first, packing finishing the day before, not the morning of. Step 2: keep each day's load honest — a few items a day beats a heroic final evening. Step 3: write the departure-day section — the walk-out list (documents in hand, phone charged, home-prep final steps) in the order they happen. Step 4: top the page with the trip, the departure date, and the days remaining, so the list orients anyone who picks it up. Step 5: keep it one page. Write {{workPath}}/report.md and call write_task_note with the report path and the heaviest day.\n\nThe deliverable `{{workPath}}/report.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}}/report.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/report.md",
+            "bytes": 1500,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "countdown",
+        "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": "Open {{workPath}}/report.md and {{workPath}}/audit.md and verify every criterion from {{workPath}}/scope.md: every audit area appears as ready or open, every itinerary-implied packing item made it onto the roster (spot-check two planned days), every open item and unpacked batch sits on a specific countdown day with lead times respected (nothing week-long scheduled for the last day), and the departure-day walk-out list is present and ordered. Write PASS/FAIL per criterion; on any failure, name the gap and loop back to countdown.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/report.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: \"countdown\" })` 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}}/report.md",
+          "artifact": true
+        }
+      ],
+      "next": "countdown"
+    },
+    {
+      "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/pr/pre-departure-countdown/versions/1.0.2/test.json

@@ -0,0 +1,100 @@
+{
+  "schemaVersion": 1,
+  "title": "Pre-Departure Countdown smoke eval",
+  "objective": "Self-contained smoke eval for the Pre-Departure Countdown craftbook using the html-page generic harness.",
+  "tags": [
+    "html-page"
+  ],
+  "prompt": "Can you biuld us a little page for this? The content notes are in source/page-content.md. Save it as index.html — one file, nothing fancy needed on our end.",
+  "setup": {
+    "projectName": "Pre-Departure Countdown Eval",
+    "about": "Self-contained eval project for pre-departure-countdown. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/index.html.",
+    "missionObjectives": "Use the Pre-Departure Countdown craftbook/template, read the seeded local fixtures, and write index.html without network calls, real credentials, or live services.",
+    "files": [
+      {
+        "path": "source/brief.md",
+        "content": "# Pre-Departure Countdown 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: pre-departure-countdown - Pre-Departure Countdown.\n"
+      },
+      {
+        "path": "source/page-content.md",
+        "content": "# Boreal Desk Support Board\n\nBuild content around a support operations board. Required labels: Open issues, Status silence, Response time, Refund leakage, Next action. Required metric text: 18 hours to 6 hours and 14.2% to 8.9%.\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Developer"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "index.html is a source-grounded self-contained HTML page/tool.",
+    "deliverables": [
+      {
+        "path": "index.html",
+        "kind": "html-page",
+        "minBytes": 3000,
+        "checks": [
+          {
+            "kind": "cssMinBytes",
+            "bytes": 650,
+            "file": "index.html"
+          },
+          {
+            "kind": "jsParses",
+            "file": "index.html"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "Boreal|Support Board|returns",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "18\\s*hours|6\\s*hours|14\\.2%|8\\.9%",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "button|input|filter|tab|card|section|nav",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "index.html",
+      "kind": "html"
+    },
+    "axes": [
+      {
+        "name": "structure",
+        "description": "Layout and hierarchy fit the content; sections are scannable and coherent."
+      },
+      {
+        "name": "grounding",
+        "description": "Seeded source content is used faithfully — no invented facts or dropped requirements."
+      },
+      {
+        "name": "interactivity",
+        "description": "Controls and states the page claims to offer actually work."
+      },
+      {
+        "name": "polish",
+        "description": "Styling is intentional and responsive rather than unstyled or broken."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "self-contained UI",
+    "source grounding",
+    "responsive structure"
+  ]
+}