@bendyline/gilde 0.1.37 → 0.1.39

package/data/craftbook-templates/oc/ocr-extract/versions/1.0.2/craftbook.json

@@ -0,0 +1,149 @@
+{
+  "id": "ocr-extract",
+  "name": "OCR a Folder of Scans",
+  "description": "Extract the text from a folder of scanned documents or photographed pages into clean, structured text — preserving reading order, paragraph breaks, and per-file provenance — then verify the extraction against the source images. Scope FIRST the languages, output structure (one file per scan vs a combined doc), and how to handle tables/columns/low-confidence regions, because OCR quality is dominated by handling layout and flagging uncertainty; then run extraction and verify a sample reads correctly. Covers OCR, text extraction, scanned documents, digitizing paper, receipt/invoice text, and image-to-text.\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 extraction (planner) — languages, output structure, layout & confidence rules → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Extract the text (developer) — OCR each scan into structured text → gated on `out` (generic-file)\n3. Verify the extraction (reviewer) — check a sample of outputs against the source images → gated on artifact `{{workPath}}/verify.md` (markdown-notes)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "scope",
+  "triggers": [
+    "ocr a folder of scans",
+    "extract text from images",
+    "digitize scanned documents",
+    "image to text",
+    "read text from photos"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the extraction",
+      "description": "languages, output structure, layout & confidence rules",
+      "prompt": "1) List the scan folder and note formats and roughly what the pages are (typed, handwritten, mixed, tables). 2) Decide the language(s) for OCR. 3) Lock the output structure: one .txt/.md per source (named to match) and/or a combined document, with each block tagged by source filename for provenance. 4) Decide layout handling — preserve paragraph breaks and reading order, how to render tables (markdown tables or labeled rows), and how to flag low-confidence or unreadable regions (e.g. [illegible]). 5) Write an acceptance-criteria checklist: 'every scan produces an output', 'reading order and paragraph breaks are preserved', 'tables are not scrambled into a single line', 'low-confidence text is flagged not silently dropped', 'output is plain readable text, not garbled'. 6) Record scope + checklist via write_task_note AND to {{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": "extract"
+    },
+    {
+      "id": "extract",
+      "name": "Extract the text",
+      "description": "OCR each scan into structured text",
+      "prompt": "1) Read scope.md. 2) For each scan, run OCR — prefer a real OCR engine (Tesseract) or the vision capability reading the image directly. 3) Preserve reading order and paragraph structure; for multi-column pages, read column-by-column not left-to-right across columns. 4) Render tables as the agreed structure rather than collapsing them. 5) Flag uncertain text with the agreed marker instead of inventing words. 6) Write each output to the agreed path with source provenance, and assemble the combined doc if requested. 7) Keep a running count of scans processed. write_task_note the output location and counts. Good looks like: text a human could read as the original document, with uncertainty honestly flagged.",
+      "suggestedRole": "developer",
+      "advanceWhen": {
+        "file": "out",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "out",
+            "bytes": 80
+          },
+          {
+            "kind": "sniff",
+            "file": "out",
+            "sniff": "nonempty"
+          }
+        ],
+        "onReject": "extract",
+        "maxAttempts": 3
+      },
+      "next": "verify"
+    },
+    {
+      "id": "verify",
+      "name": "Verify the extraction",
+      "description": "check a sample of outputs against the source images",
+      "prompt": "1) Re-read scope.md's checklist. 2) Confirm output count equals scan count (PASS/FAIL). 3) Pick 3-4 scans, open BOTH the image and its extracted text, and compare: is the text accurate, in correct reading order, with paragraphs/tables intact (PASS/FAIL)? 4) Confirm low-confidence regions are flagged, not silently wrong or dropped. 5) Confirm provenance — each block traces to its source file. 6) Write a verification note with PASS/FAIL per criterion and the specific files you checked. On a loop-back the extractor fixes the flagged files. write_task_note the result.\n\nThe deliverable `{{workPath}}/verify.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",
+      "consumes": [
+        {
+          "file": "out"
+        }
+      ],
+      "advanceWhen": {
+        "file": "{{workPath}}/verify.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "out",
+            "bytes": 80
+          },
+          {
+            "kind": "sniff",
+            "file": "out",
+            "sniff": "nonempty"
+          }
+        ],
+        "onReject": "verify",
+        "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": "1) Open out/ and confirm every scan has an output (PASS/FAIL). 2) For 3-4 files, open the source image alongside the extracted text and check accuracy, reading order, and that tables/columns are not scrambled (PASS/FAIL). 3) Confirm uncertain text is flagged, not silently dropped or hallucinated. 4) Confirm provenance back to source files. 5) Reconcile verify.md's claims with what you observe. 6) Write PASS/FAIL per criterion. Inaccurate/garbled extraction → loop back to extract.\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: \"extract\" })` 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": "out"
+        }
+      ],
+      "next": "extract"
+    },
+    {
+      "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/oc/ocr-extract/versions/1.0.2/test.json

@@ -0,0 +1,106 @@
+{
+  "schemaVersion": 1,
+  "title": "OCR a Folder of Scans smoke eval",
+  "objective": "Self-contained smoke eval for the OCR a Folder of Scans 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": "OCR a Folder of Scans Eval",
+    "about": "Self-contained eval project for ocr-extract. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/media-plan.json.",
+    "missionObjectives": "Use the OCR a Folder of Scans 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": "# OCR a Folder of Scans 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: ocr-extract - OCR a Folder of Scans.\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": "ocr-extract-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/of/offer-compare/versions/1.0.2/craftbook.json

@@ -0,0 +1,118 @@
+{
+  "id": "offer-compare",
+  "name": "Offer Comparison",
+  "description": "Turn competing job offers into one honest decision sheet — every offer normalized into the same fact grid (comp, equity with vesting caveats, benefits, level, location, growth, deadlines), unknowns marked UNKNOWN instead of guessed, computed year-1 and 4-year views under stated assumptions, a weighted read against the candidate's OWN priorities, and a recommendation that separates arithmetic from judgment. Use when any second offer (or a strong counter) lands.\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. Normalize the facts (planner) — one fact grid across every offer, unknowns honest, the user's own weights elicited → gated on artifact `{{workPath}}/offer-facts.md` (markdown-notes)\n2. Build the comparison (planner) — the decision sheet — table, computed views, weighted read, recommendation → gated on artifact `{{workPath}}/reports/offer-compare.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": "facts",
+  "triggers": [
+    "compare my offers",
+    "which offer should I take",
+    "offer comparison",
+    "help me decide between offers"
+  ],
+  "steps": [
+    {
+      "id": "facts",
+      "name": "Normalize the facts",
+      "description": "one fact grid across every offer, unknowns honest, the user's own weights elicited",
+      "prompt": "Build the fact sheet before any comparison. Step 1: collect every offer's terms from the user or project documents into one grid: base, bonus (target and realism), equity (kind, vesting schedule, valuation caveats), benefits, PTO, level/title, location/remote terms, team and manager, growth story, response deadline. Step 2: mark every unknown as UNKNOWN — never estimate a number the offer letter does not state; list the UNKNOWNs as questions to ask each company. Step 3: elicit the user's OWN decision weights — what actually matters to them (comp, growth, stability, location, mission, team) — do not invent or assume weights. Step 4: state the assumptions any computed view will use (equity valuation basis, bonus attainment). Step 5: write an acceptance-criteria checklist ('every offer covered on every factor', 'unknowns marked UNKNOWN', 'weights are the user's own words', 'assumptions stated'). Write the fact sheet + checklist via write_task_note AND to {{workPath}}/offer-facts.md.\n\nThe deliverable `{{workPath}}/offer-facts.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}}/offer-facts.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/offer-facts.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/offer-facts.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "facts",
+        "maxAttempts": 3
+      },
+      "next": "compare"
+    },
+    {
+      "id": "compare",
+      "name": "Build the comparison",
+      "description": "the decision sheet — table, computed views, weighted read, recommendation",
+      "prompt": "Write {{workPath}}/reports/offer-compare.md from the fact sheet. Step 1: a factor-by-offer markdown comparison table — offers as columns, factors as rows, UNKNOWNs still visible. Step 2: computed lines under the STATED assumptions: year-1 cash, and a 4-year total-comp view with the equity caveats repeated next to the number. Step 3: a weighted read against the user's stated priorities — show how each offer scores on what THEY said matters, not on a generic rubric. Step 4: risks and negotiation levers per offer (deadline pressure, level mismatch, unvested cliff, remote-policy risk). Step 5: a recommendation that separates the arithmetic ('offer B leads on 4-year comp under these assumptions') from the judgment ('offer A fits the growth priority you weighted highest') — and says plainly which way the sheet leans and why. On a loop-back, fix ONLY the named issues.\n\nThe deliverable `{{workPath}}/reports/offer-compare.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}}/reports/offer-compare.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/reports/offer-compare.md",
+            "bytes": 1500,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/reports/offer-compare.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "compare",
+        "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 decision sheet. Step 1: recompute the year-1 and 4-year lines from the fact sheet — FAIL on any arithmetic that does not reproduce, or any computed number whose assumption is not stated beside it. Step 2: check every offer appears in every table row and that fact-sheet UNKNOWNs were not silently replaced with guesses. Step 3: trace the weighted read to the user's stated priorities — FAIL a rubric the user never gave. Step 4: confirm the recommendation separates arithmetic from judgment and takes a clear position. Write PASS/FAIL per criterion.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/reports/offer-compare.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: \"compare\" })` 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}}/reports/offer-compare.md",
+          "artifact": true
+        }
+      ],
+      "next": "compare"
+    },
+    {
+      "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/of/offer-compare/versions/1.0.2/test.json

@@ -0,0 +1,113 @@
+{
+  "schemaVersion": 1,
+  "title": "Offer Comparison smoke eval",
+  "objective": "Self-contained smoke eval for the Offer Comparison craftbook using the corpus generic harness.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "We need a offer comparison for the returns-desk work. The backgound info is in source/brief.md — have a look and write it up in report.md.",
+  "setup": {
+    "projectName": "Offer Comparison Eval",
+    "about": "Self-contained eval project for offer-compare. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/report.md.",
+    "missionObjectives": "Use the Offer Comparison craftbook/template, read the seeded local fixtures, and write report.md without network calls, real credentials, or live services.",
+    "files": [
+      {
+        "path": "source/brief.md",
+        "content": "# Offer Comparison 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: offer-compare - Offer Comparison.\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Writer"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "report.md is a source-grounded markdown artifact with required facts and actions.",
+    "deliverables": [
+      {
+        "path": "report.md",
+        "kind": "markdown-report",
+        "minBytes": 1000,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "source/brief\\.md|Brief",
+            "flags": "i",
+            "label": "cite or name the seeded source"
+          },
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "18\\b[\\s\\S]{0,30}\\bSKUs",
+            "flags": "i",
+            "label": "include the 18-SKU pilot scope"
+          },
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "18\\s*hours|6\\s*hours",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "14\\.2%|8\\.9%",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "status silence|photo submission",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "report.md",
+            "pattern": "automated status emails|barcode|Finance",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "report.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "grounding",
+        "description": "Seeded facts appear faithfully and the source is cited."
+      },
+      {
+        "name": "structure",
+        "description": "Sections progress logically for the artifact class."
+      },
+      {
+        "name": "completeness",
+        "description": "Required elements (risks, next actions, scope) are all present."
+      },
+      {
+        "name": "tone",
+        "description": "The register fits the audience and artifact class."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "source-grounded prose",
+    "section structure",
+    "next-action clarity"
+  ]
+}

package/data/craftbook-templates/on/onboarding-checklist/versions/1.0.2/craftbook.json

@@ -0,0 +1,119 @@
+{
+  "id": "onboarding-checklist",
+  "name": "Generate an Onboarding Checklist",
+  "description": "Produce a clear, role-specific onboarding checklist for a new hire, customer, or user — grouped by phase (before day one, week one, first 30 days), with each item owned, actionable, and verifiable. Scopes the audience, the phases, and a 'good item' rule FIRST so the checklist is complete and non-overlapping, then writes the actual checklist as a clean markdown document with grouped, owner-tagged, checkbox items. Scoping the phases and item-quality bar before writing is what turns a vague to-do dump into a sequenced, accountable onboarding plan.\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 checklist (planner) — audience, phases, owners, and the good-item rule → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Write the checklist (copywriter) — produce the grouped, owner-tagged checklist doc → gated on `onboarding-checklist.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": "scope",
+  "triggers": [
+    "create an onboarding checklist",
+    "new hire checklist",
+    "customer onboarding steps",
+    "onboarding plan",
+    "first 30 days checklist"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the checklist",
+      "description": "audience, phases, owners, and the good-item rule",
+      "prompt": "Define the checklist before writing it. Step 1: pin the audience and context (which role/customer type, team or product, and any tools/accounts/access they will need). Step 2: choose the phase grouping (e.g. Before Day One, Day One, Week One, First 30 Days) and what each phase is meant to achieve. Step 3: define the owner set (who is responsible for each kind of item — IT, manager, the person themselves, buddy) so every item can be tagged. Step 4: define the 'good item' rule — each item is a single concrete action, has an owner, has a done-signal (how you know it's complete), and items do not overlap or leave gaps. Step 5: write an acceptance-criteria checklist ('grouped into the chosen phases in order', 'every item is one actionable step with an owner', 'every item has a verifiable done-signal', 'no duplicate or overlapping items', 'access/tooling and a day-one welcome are covered', 'rendered as checkbox items'). Call write_task_note with the audience + phases + owners + rule + checklist and write {{workPath}}/scope.md. No checklist content yet.\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": "write"
+    },
+    {
+      "id": "write",
+      "name": "Write the checklist",
+      "description": "produce the grouped, owner-tagged checklist doc",
+      "prompt": "Write the onboarding checklist to the locked scope. Step 1: open with a one-line purpose and who it's for. Step 2: create a section header per phase in order. Step 3: under each phase, write items as markdown checkboxes ('- [ ] '), each a single concrete action, prefixed or suffixed with its owner (e.g. '(IT)' / '(Manager)' / '(You)') and a short done-signal where it adds clarity. Step 4: ensure coverage — accounts/access provisioning, equipment, key intro meetings, required training, a day-one welcome, and 30-day check-in are all present, with no duplicate or overlapping items. Step 5: keep wording tight and imperative ('Send', 'Grant', 'Schedule'); write the document to the deliverable path. Call write_task_note with the document path and the item count per phase.",
+      "suggestedRole": "copywriter",
+      "advanceWhen": {
+        "file": "onboarding-checklist.md",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "onboarding-checklist.md",
+            "bytes": 500
+          }
+        ],
+        "scripts": [
+          {
+            "name": "checkContains",
+            "scope": "standard",
+            "inputs": {
+              "file": "onboarding-checklist.md",
+              "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+              "flags": "i"
+            }
+          }
+        ],
+        "onReject": "write",
+        "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 onboarding-checklist.md and verify every criterion from {{workPath}}/scope.md: items are grouped into the chosen phases in order, every item is a single actionable step rendered as a checkbox with an owner tag and a verifiable done-signal, there are no duplicate or overlapping items, and coverage includes access/tooling, a day-one welcome, key meetings, training, and a 30-day check-in. Write PASS/FAIL per criterion; on any failure, name the gap and loop back to write.\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: \"write\" })` 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": "onboarding-checklist.md"
+        }
+      ],
+      "next": "write"
+    },
+    {
+      "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"
+}