@bendyline/gilde 0.1.38 → 0.1.39

package/data/craftbook-templates/vi/visit-prep/versions/1.0.2/craftbook.json

@@ -0,0 +1,152 @@
+{
+  "id": "visit-prep",
+  "name": "Care Visit Prep",
+  "description": "Prepare a one-page brief for a medical or care appointment from the project's care record: compile everything logged since the last visit, the current medication list with doses, the changes and symptoms worth raising, and a prioritized question list — so the ten minutes in the room get spent on what matters. Locks the target appointment and the brief's sections FIRST, then compiles from the record, then writes the brief. Use for doctor visit prep, specialist appointments, care-team check-ins, and 'what do we need to tell the doctor'.\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 visit (planner) — target appointment, brief sections, what counts as raise-worthy → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Compile the record (data-analyst) — window events, med list, changes, draft questions → gated on artifact `{{workPath}}/compile.md` (markdown-notes)\n3. Write the brief (copywriter) — one-page visit brief, questions first → 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": [
+    "prep for the appointment",
+    "visit prep",
+    "what do we tell the doctor",
+    "prepare for the doctor",
+    "appointment brief"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the visit",
+      "description": "target appointment, brief sections, what counts as raise-worthy",
+      "prompt": "Define the prep before compiling. Step 1: read care.json and meds.json from the workspace and identify the target appointment (the next upcoming one, or the one the owner names) and the last visit before it — the window between them is the compile range. Step 2: fix the brief's sections — current medications with doses; changes since last visit (meds started/stopped, new symptoms, incidents); patterns worth raising (sleep, appetite, mood, falls); the question list; and logistics (who attends, what to bring). Step 3: agree with the owner what counts as raise-worthy versus noise so the brief stays one page. Step 4: write an acceptance-criteria checklist ('every appointment-window event considered', 'med list matches the active roster exactly', 'each raised item cites the logged entry behind it', 'questions ranked, most important first', 'one page'). Call write_task_note with the appointment + window + 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": "compile"
+    },
+    {
+      "id": "compile",
+      "name": "Compile the record",
+      "description": "window events, med list, changes, draft questions",
+      "prompt": "Compile the brief's raw material from the record. Step 1: pull every care-log entry in the window and sort it into the agreed sections — changes, symptoms, patterns, noise. Step 2: pull the active medication list with doses from the meds roster and note anything started, stopped, or changed in the window. Step 3: for each raise-worthy item, attach the dates and the logged entries behind it so the clinician gets specifics, not vibes. Step 4: draft the question list from the raised items plus anything the owner adds, and rank it most-important-first. Step 5: write the compiled material to {{workPath}}/compile.md and call write_task_note with the counts — window events, meds, raised items, questions.\n\nThe deliverable `{{workPath}}/compile.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": "data-analyst",
+      "advanceWhen": {
+        "file": "{{workPath}}/compile.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/compile.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/compile.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "compile",
+        "maxAttempts": 3
+      },
+      "next": "brief"
+    },
+    {
+      "id": "brief",
+      "name": "Write the brief",
+      "description": "one-page visit brief, questions first",
+      "prompt": "Write the one-page visit brief from the compiled material. Step 1: header — who the visit is for, the appointment, who attends. Step 2: the ranked questions, numbered, at the top — the room time goes here first. Step 3: current medications with doses, flagging window changes. Step 4: since-last-visit summary — the raised items with their dates, briefest sufficient wording. Step 5: close with logistics and space for the clinician's answers. One page, printable, calm. Write {{workPath}}/report.md and call write_task_note with the report path and the top question.\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": "brief",
+        "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}}/compile.md and verify every criterion from {{workPath}}/scope.md: the medication list matches the active meds roster exactly (check every med and dose), every raised item traces to a logged entry with a date, every window event was sorted or consciously set aside, the questions are ranked with the most important first, and the brief fits one page. Write PASS/FAIL per criterion; on any failure, name the gap and loop back to brief.\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: \"brief\" })` 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": "brief"
+    },
+    {
+      "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/vi/visit-prep/versions/1.0.2/test.json

@@ -0,0 +1,100 @@
+{
+  "schemaVersion": 1,
+  "title": "Care Visit Prep smoke eval",
+  "objective": "Self-contained smoke eval for the Care Visit Prep 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": "Care Visit Prep Eval",
+    "about": "Self-contained eval project for visit-prep. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/index.html.",
+    "missionObjectives": "Use the Care Visit Prep 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": "# Care Visit Prep 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: visit-prep - Care Visit Prep.\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"
+  ]
+}

package/data/craftbook-templates/vo/voiceover-script/versions/1.0.2/craftbook.json

@@ -0,0 +1,119 @@
+{
+  "id": "voiceover-script",
+  "name": "Voiceover / TTS Script",
+  "description": "Write a voiceover script tuned for spoken delivery — by a human VO artist or a text-to-speech engine — for a video, ad, explainer, or narration, with pacing, pronunciation, and timing baked in. Briefs the audience, runtime, and voice FIRST, then drafts a performance-ready script with scene/segment breaks, [pause] and emphasis cues, phonetic spellings for tricky words, and a per-segment word count that fits the target seconds. The brief-then-draft ordering ensures the script is written for the EAR and the clock, not the page — short sentences, read-aloud rhythm, and a runtime that actually fits.\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. Voiceover brief (copywriter) — audience, runtime, voice/tone, pronunciation list → gated on artifact `{{workPath}}/brief.md` (markdown-notes)\n2. Draft the script (copywriter) — performance-ready script with cues, pacing, timing → gated on `voiceover-script.md` (markdown-doc)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "brief",
+  "triggers": [
+    "write a voiceover script",
+    "narration script",
+    "tts script",
+    "script for a video",
+    "explainer voiceover"
+  ],
+  "steps": [
+    {
+      "id": "brief",
+      "name": "Voiceover brief",
+      "description": "audience, runtime, voice/tone, pronunciation list",
+      "prompt": "Lock the brief BEFORE drafting. Step 1: capture the purpose, audience, and the single takeaway the listener must leave with. Step 2: set the target runtime in seconds and translate it to a word budget (~150 words/minute for natural narration; record the budget). Step 3: define the voice/tone (warm, authoritative, energetic) and the delivery (single narrator vs dialogue). Step 4: list tricky names/terms/numbers with their intended pronunciation (phonetic respelling) and how numbers/units should be spoken. Step 5: decide the segment structure (hook -> body beats -> CTA). Write an acceptance-criteria checklist (total words within ±10% of the budget, sentences short and read-aloud friendly, every segment has a target second count, pause/emphasis cues present, tricky words have phonetics, ends on the CTA) and decisions to `{{workPath}}/brief.md` via write_task_note AND the produces path.\n\nThe deliverable `{{workPath}}/brief.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}}/brief.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/brief.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/brief.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "brief",
+        "maxAttempts": 3
+      },
+      "next": "draft"
+    },
+    {
+      "id": "draft",
+      "name": "Draft the script",
+      "description": "performance-ready script with cues, pacing, timing",
+      "prompt": "Write the performance-ready script to the brief. Step 1: structure it into labeled segments (e.g. `[0:00-0:08] HOOK`) each with a target second count and an inline running word count. Step 2: write for the ear — short declarative sentences, active voice, one idea per line, conversational rhythm; read each line aloud in your head and cut anything that trips the tongue. Step 3: add performance cues inline: `[pause]`, `[beat]`, emphasis via *asterisks* or CAPS-for-stress, and tone notes per segment. Step 4: use the brief's phonetic respellings for tricky words and write numbers/units as they should be SPOKEN (e.g. 'twenty twenty-six', 'three hundred dollars'). Step 5: end on the locked CTA. Write the full script to `voiceover-script.md` with the segment headers and a final total word count + estimated runtime. On a loop-back, fix only the flagged segments (over-budget, hard-to-say, missing cues).",
+      "suggestedRole": "copywriter",
+      "advanceWhen": {
+        "file": "voiceover-script.md",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "voiceover-script.md",
+            "bytes": 500
+          }
+        ],
+        "scripts": [
+          {
+            "name": "checkContains",
+            "scope": "standard",
+            "inputs": {
+              "file": "voiceover-script.md",
+              "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+              "flags": "i"
+            }
+          }
+        ],
+        "onReject": "draft",
+        "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": "Read it aloud and time it. Step 1: confirm total word count is within ±10% of the brief's budget and the stated runtime matches at ~150 wpm (PASS/FAIL). Step 2: confirm each segment is labeled with a target second count and the running counts add up. Step 3: read several lines aloud — are sentences short and natural, with no tongue-twisters? Step 4: confirm pause/emphasis cues are present and tricky words use the phonetic respellings. Step 5: confirm numbers are written spoken-style and it ends on the CTA. 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: \"draft\" })` 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": "voiceover-script.md"
+        }
+      ],
+      "next": "draft"
+    },
+    {
+      "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/vo/voiceover-script/versions/1.0.2/test.json

@@ -0,0 +1,106 @@
+{
+  "schemaVersion": 1,
+  "title": "Voiceover / TTS Script smoke eval",
+  "objective": "Self-contained smoke eval for the Voiceover / TTS Script 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": "Voiceover / TTS Script Eval",
+    "about": "Self-contained eval project for voiceover-script. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/media-plan.json.",
+    "missionObjectives": "Use the Voiceover / TTS Script 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": "# Voiceover / TTS Script 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: voiceover-script - Voiceover / TTS Script.\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": "voiceover-script-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/we/weak-spot-drill/versions/1.0.2/craftbook.json

@@ -0,0 +1,166 @@
+{
+  "id": "weak-spot-drill",
+  "name": "Weak-Spot Drill",
+  "description": "Stop practicing what you already know: read the review record, name the actual weak spots with the numbers as evidence, and build a short targeted drill plus the follow-up schedule that closes them. Locks the ranked weak list FIRST — from the deck's boxes and miss counts, not vibes — then builds the drill page around only that material, then writes the plan that says when the spot counts as closed. Use for targeted practice, 'I keep missing the same ones', closing weak spots, drilling trouble topics, and study sessions that skip what already sticks.\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. Rank the weak spots (studiemaat) — weak spots named from the record, with the numbers as evidence → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Build the drill (developer) — short interactive drill.html over only the weak material → gated on `drill.html` (html-page)\n3. Write the follow-up plan (coach) — the schedule that closes each weak spot → 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": [
+    "drill my weak spots",
+    "i keep missing the same ones",
+    "targeted practice from the results",
+    "what am i weakest on",
+    "make me a drill"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Rank the weak spots",
+      "description": "weak spots named from the record, with the numbers as evidence",
+      "prompt": "Find the real weak spots before building anything. Step 1: read deck.json from the workspace — every card's box and its got-it/missed stats are the review record. Step 2: rank the weakness with evidence — cards stuck in box 1, cards missed as often as got, and the themes those cards cluster into; write each weak spot with its numbers ('irregular verbs: 5 of 6 cards in box 1, 11 misses against 4 got-its'), never a hunch. Step 3: lock the drill rules — only ranked weak material goes in, ten to fifteen items, each item coming at the card's idea from a fresh angle rather than repeating the card verbatim. Step 4: write an acceptance-criteria checklist ('every drill item traces to a ranked weak card', 'no item drills material the record says is strong', 'the drill checks answers on the spot', 'the follow-up plan says when each spot counts as closed'). Call write_task_note with the ranked list + 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": "studiemaat",
+      "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": "drill"
+    },
+    {
+      "id": "drill",
+      "name": "Build the drill",
+      "description": "short interactive drill.html over only the weak material",
+      "prompt": "Build the drill as one self-contained page at drill.html. Step 1: write the items from the ranked weak spots only — a strong card in the drill is a miss against the scope; come at each weak idea from a new angle (reversed prompt, a fill-in, a which-of-these) instead of re-showing the card. Step 2: make it interactive — the taker answers item by item, gets told right or wrong on the spot with the correct answer shown, and sees a simple tally at the end. Step 3: keep the tone warm — a drill on weak material stings enough already; the feedback should coach, not scold. Step 4: one file, inline styles and script, and check the script parses before completing. Call write_task_note with the item count and which weak spots each item targets.",
+      "suggestedRole": "developer",
+      "advanceWhen": {
+        "file": "drill.html",
+        "minBytes": 1,
+        "sniff": "html-complete"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "drill.html",
+            "bytes": 2000
+          },
+          {
+            "kind": "htmlLint",
+            "file": "drill.html"
+          },
+          {
+            "kind": "cssMinBytes",
+            "bytes": 400,
+            "file": "drill.html"
+          },
+          {
+            "kind": "jsParses",
+            "file": "drill.html"
+          }
+        ],
+        "scripts": [
+          {
+            "name": "checkHtmlComplete",
+            "scope": "standard",
+            "inputs": {
+              "file": "drill.html"
+            }
+          }
+        ],
+        "onReject": "drill",
+        "maxAttempts": 3
+      },
+      "next": "plan"
+    },
+    {
+      "id": "plan",
+      "name": "Write the follow-up plan",
+      "description": "the schedule that closes each weak spot",
+      "prompt": "Write the follow-up at {{workPath}}/report.md. Step 1: restate the ranked weak spots with their numbers — the report should make sense to someone who never saw the scope notes. Step 2: give the schedule — when to run the drill again, which cards to put back into review through the project's study capabilities, and the spacing between passes. Step 3: define closed — the concrete signal (a clean drill run, the cards climbing out of box 1) that says a spot no longer needs drilling. Step 4: end with one encouraging, honest line about the trend. Keep it under a page. Write {{workPath}}/report.md and call write_task_note with the report path and the next review date.\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": 800,
+            "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": "plan",
+        "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 drill.html, {{workPath}}/report.md, and deck.json and verify every criterion from {{workPath}}/scope.md: every drill item traces to a ranked weak card — spot-check each against the record, an item drilling strong material means the drill is padding, and a generic quiz over the whole deck is the classic failure; the drill checks answers on the spot; the report restates the weak spots with their numbers, schedules the re-runs and re-reviews, and defines when each spot counts as closed. Write PASS/FAIL per criterion; on any failure, name the item or spot and loop back to drill.\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: \"drill\" })` 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": "drill"
+    },
+    {
+      "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"
+}