@bendyline/gilde 0.1.37 → 0.1.39

package/data/craftbook-templates/da/data-quality-audit/versions/1.0.2/test.json

@@ -0,0 +1,329 @@
+{
+  "schemaVersion": 1,
+  "title": "Data quality scorecard over seeded customer CSV",
+  "objective": "Measure whether the data-quality-audit craftbook defines measurable rules, emits audit JSON, and reports score/remediation with examples.",
+  "tags": [
+    "data"
+  ],
+  "prompt": "In the `Data Quality Audit Eval` project, read `data/customers.csv` and `data/quality-rules.md`, then write `data/audit.json` and `tasks/eval/report.md`. Apply the seeded rules exactly: audit date, required fields, allowed statuses, email format, date consistency, duplicate customer IDs, and stale last_seen threshold. The JSON must include summary.total_records equal to the CSV data row count, per-dimension pass_rate values, failing_examples, and overall_score. The report must cover completeness, validity, consistency, uniqueness, timeliness, overall score, failing examples, and prioritized remediation.",
+  "setup": {
+    "projectName": "Data Quality Audit Eval",
+    "about": "Self-contained eval project for data-quality-audit. Source data is workspace/data/customers.csv and deterministic audit rules are in workspace/data/quality-rules.md.",
+    "missionObjectives": "Produce a measured data-quality scorecard using the seeded rule file, with JSON audit data, per-dimension pass rates, failing examples, overall score, and remediation plan.",
+    "files": [
+      {
+        "path": "data/customers.csv",
+        "content": "customer_id,email,status,start_date,end_date,last_seen\nC-001,ana@example.com,active,2026-01-10,,2026-06-20\nC-002,,trial,2026-02-02,,2026-02-02\nC-003,bad-email,active,2026-03-01,,2026-06-18\nC-004,lee@example.com,churned,2026-01-15,2025-12-31,2026-01-20\nC-004,lee.dup@example.com,active,2026-01-15,,2026-06-12\nC-005,maya@example.com,pending,2026-04-01,,2026-04-02\n"
+      },
+      {
+        "path": "data/quality-rules.md",
+        "content": "# Data quality rules\n\nAudit as of: 2026-06-28.\n\nRequired fields: customer_id, email, status, start_date, last_seen.\nAllowed status values: active, trial, churned. Treat any other status as invalid.\nEmail validity: an email must contain exactly one @ and a dotted domain.\nDate consistency: if end_date is present, it must be on or after start_date.\nUniqueness: customer_id must be unique across rows.\nTimeliness: active or trial records are stale when last_seen is more than 60 days before the audit date. Invalid-status records should still be assessed for timeliness.\nOverall score: report the percentage of rule checks that passed and name it overall_score in data/audit.json.\n\nExpected minimum failing examples to find: C-002 missing email and stale last_seen; C-003 invalid email; C-004 duplicate customer_id and impossible end_date; C-005 invalid status and stale last_seen.\n"
+      }
+    ],
+    "worker": {
+      "name": "Oda",
+      "role": "Data Analyst"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "workspace/report.md and data/audit.json apply the seeded audit rules with per-dimension pass rates, failing examples, overall score, and remediation.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/report.md",
+        "kind": "markdown-report",
+        "minBytes": 2200,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "completeness",
+            "flags": "i",
+            "label": "include a completeness section/result"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "validity",
+            "flags": "i",
+            "label": "include a validity section/result"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "consistency",
+            "flags": "i",
+            "label": "include a consistency section/result"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "uniqueness",
+            "flags": "i",
+            "label": "include a uniqueness section/result"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "timeliness|stale|last_seen",
+            "flags": "i",
+            "label": "include timeliness results"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "2026-06-28",
+            "flags": "i",
+            "label": "include the seeded audit date"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "6\\s+records|Total Records\\s*\\|\\s*6",
+            "flags": "i",
+            "label": "use the six CSV rows"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "C-002",
+            "flags": "i",
+            "label": "include C-002 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "C-003",
+            "flags": "i",
+            "label": "include C-003 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "C-004",
+            "flags": "i",
+            "label": "include C-004 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "C-005",
+            "flags": "i",
+            "label": "include C-005 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "C-005[\\s\\S]{0,220}(pending|invalid status|status)|pending[\\s\\S]{0,220}C-005",
+            "flags": "i",
+            "label": "include C-005 invalid pending status, not only stale last_seen"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "pass rate|pass_rate",
+            "flags": "i",
+            "label": "include pass rates"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "overall(?:\\s+data\\s+quality)?\\s+score|overall_score",
+            "flags": "i",
+            "label": "include an explicit overall score"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "failing examples?",
+            "flags": "i",
+            "label": "include a failing examples section"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "remediation|priority|fix",
+            "flags": "i"
+          },
+          {
+            "kind": "notContains",
+            "file": "tasks/eval/report.md",
+            "pattern": "first_name|last_name|john@doe|90[- ]day|90\\s+days|2024-\\d\\d-\\d\\d|\\b7\\s+records\\b|\\d\\s*/\\s*7\\b",
+            "flags": "i",
+            "label": "avoid invented columns, values, thresholds, and audit dates not in the seeded rules"
+          }
+        ]
+      },
+      {
+        "path": "data/audit.json",
+        "kind": "json",
+        "minBytes": 700,
+        "checks": [
+          {
+            "kind": "sniff",
+            "file": "data/audit.json",
+            "sniff": "json-valid"
+          },
+          {
+            "kind": "jsonPathEquals",
+            "file": "data/audit.json",
+            "path": "audit_date",
+            "value": "2026-06-28",
+            "label": "audit_date must equal the seeded audit date from quality-rules.md"
+          },
+          {
+            "kind": "jsonPathEquals",
+            "file": "data/audit.json",
+            "path": "summary.total_records",
+            "value": 6,
+            "label": "summary.total_records must equal the six CSV data rows"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "2026-06-28",
+            "flags": "i",
+            "label": "include the seeded audit date"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "overall_score|overallScore|overall[^\"]*score",
+            "flags": "i",
+            "label": "include overall_score"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "pass_rate|passRate",
+            "flags": "i",
+            "label": "include per-dimension pass rates"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "failing_examples|failingExamples",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "completeness",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "validity",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "consistency",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "uniqueness",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "timeliness|stale|last_seen",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "C-002",
+            "flags": "i",
+            "label": "include C-002 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "C-003",
+            "flags": "i",
+            "label": "include C-003 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "C-004",
+            "flags": "i",
+            "label": "include C-004 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "C-005",
+            "flags": "i",
+            "label": "include C-005 as a failing example"
+          },
+          {
+            "kind": "contains",
+            "file": "data/audit.json",
+            "pattern": "C-005[\\s\\S]{0,220}(pending|invalid status|status)|pending[\\s\\S]{0,220}C-005",
+            "flags": "i",
+            "label": "include C-005 invalid pending status, not only stale last_seen"
+          },
+          {
+            "kind": "notContains",
+            "file": "data/audit.json",
+            "pattern": "first_name|last_name|john@doe|90[- ]day|90\\s+days|\"overall_score\"\\s*:\\s*\"[A-F][+-]?\"",
+            "flags": "i",
+            "label": "avoid invented columns, values, thresholds, and letter-grade overall scores"
+          },
+          {
+            "kind": "notContains",
+            "file": "data/audit.json",
+            "pattern": "2024-\\d\\d-\\d\\d",
+            "flags": "i",
+            "label": "do not invent an unrelated audit timestamp"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/report.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "accuracy",
+        "description": "Computed figures are correct against the seeded records."
+      },
+      {
+        "name": "grounding",
+        "description": "Every claim traces to the seeded data — nothing invented."
+      },
+      {
+        "name": "structure",
+        "description": "Tables and sections make the analysis checkable at a glance."
+      },
+      {
+        "name": "actionability",
+        "description": "Recommendations follow from the findings and are concretely doable."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "measurable rules",
+    "failing examples",
+    "prioritized remediation"
+  ]
+}

package/data/craftbook-templates/da/data-to-report/versions/1.0.2/craftbook.json

@@ -0,0 +1,152 @@
+{
+  "id": "data-to-report",
+  "name": "Dataset to Narrative Report",
+  "description": "Turn a dataset into a written, evidence-backed analytical report that a non-technical reader can act on. Analyzes the data FIRST to surface the real findings (key metrics, segments, trends, surprises) with supporting numbers, then writes a narrative that leads with the takeaway and traces every claim to a figure, then builds the polished report document with sections, a summary table, and an explicit limitations note. Use this for turning raw numbers into an insights report, executive summary, data storytelling, or a findings write-up stakeholders will read.\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. Analyze the data (data-analyst) — extract the real findings with supporting numbers → gated on artifact `{{workPath}}/analyze.md` (markdown-notes)\n2. Write the narrative (copywriter) — draft a takeaway-led narrative grounded in the figures → gated on artifact `{{workPath}}/write.md` (markdown-notes)\n3. Build the report (developer) — assemble the polished {{workPath}}/report.md with sections and a summary table → 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": "analyze",
+  "triggers": [
+    "turn data into a report",
+    "analyze this dataset",
+    "write a findings report",
+    "data insights writeup",
+    "executive summary from data"
+  ],
+  "steps": [
+    {
+      "id": "analyze",
+      "name": "Analyze the data",
+      "description": "extract the real findings with supporting numbers",
+      "prompt": "Interrogate the dataset to find what actually matters — do not write prose yet. Step 1: state the 3-5 questions the report should answer (totals, trends over time, top/bottom segments, notable correlations or anomalies). Step 2: compute the answer to each with concrete numbers (counts, means, %, deltas, top-N) and record the exact figure plus how you derived it. Step 3: identify the single most important takeaway and 2-3 supporting findings, each tied to a number. Step 4: note data caveats (sample size, missing periods, definitions). Step 5: write an acceptance-criteria checklist for the report ('leads with the headline finding', 'every claim cites a number', 'includes a summary table', 'states limitations'). Call write_task_note and write the findings + figures + checklist to {{workPath}}/analyze.md. No narrative prose yet.\n\nThe deliverable `{{workPath}}/analyze.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}}/analyze.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/analyze.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/analyze.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "analyze",
+        "maxAttempts": 3
+      },
+      "next": "write"
+    },
+    {
+      "id": "write",
+      "name": "Write the narrative",
+      "description": "draft a takeaway-led narrative grounded in the figures",
+      "prompt": "Write the report narrative using ONLY the figures from the analysis note — invent no numbers. Step 1: open with a 2-3 sentence executive summary that states the headline finding and its 'so what'. Step 2: write one short section per supporting finding; each paragraph states the insight, then the number that proves it, then the implication. Step 3: keep it scannable — short paragraphs, plain language, no jargon a manager wouldn't use. Step 4: end with a clearly labeled 'Limitations' paragraph from the caveats. Call write_task_note with the draft keyed by section so the build phase drops it straight in.\n\nThe deliverable `{{workPath}}/write.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}}/write.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/write.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/write.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "write",
+        "maxAttempts": 3
+      },
+      "next": "build"
+    },
+    {
+      "id": "build",
+      "name": "Build the report",
+      "description": "assemble the polished {{workPath}}/report.md with sections and a summary table",
+      "prompt": "Assemble the final report as {{workPath}}/report.md. Step 1: add a title, a date, and an Executive Summary section from the approved narrative. Step 2: add one ## section per finding with the prose and inline figures. Step 3: include a Markdown summary table of the key metrics (metric, value, change). Step 4: add a Methodology / Data notes section and a Limitations section. Step 5: ensure every numeric claim in the prose matches the analysis note. On a loop-back, fix only the named gaps. Call write_task_note with the report path and section list.\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": "developer",
+      "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": "build",
+        "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 {{workPath}}/report.md as a stakeholder would. Verify: it opens with a takeaway-led executive summary; every numeric claim matches a figure in the analysis note (re-check at least three); it contains a summary table of key metrics; section headers organize it; and a Limitations section is present and honest. Confirm no fabricated numbers and no claim without a supporting figure. Write PASS/FAIL per acceptance criterion; any FAIL loops back to build.\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: \"build\" })` 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": "build"
+    },
+    {
+      "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/da/data-to-report/versions/1.0.2/test.json

@@ -0,0 +1,98 @@
+{
+  "schemaVersion": 1,
+  "title": "Executive report from seeded data",
+  "objective": "Measure whether the data-to-report craftbook analyzes seeded data first and produces a narrative report grounded in computed figures.",
+  "tags": [
+    "data"
+  ],
+  "prompt": "In the `Data to Report Eval` project, read `data/returns-summary.csv` and write `tasks/eval/report.md`: an executive report for support leadership. Include an executive summary, key metrics table, findings, segment trends, recommendations, and limitations. Ground every claim in the seeded figures.",
+  "setup": {
+    "projectName": "Data to Report Eval",
+    "about": "Self-contained eval project for the data-to-report craftbook. The dataset is workspace/data/returns-summary.csv.",
+    "missionObjectives": "Write an executive analytical report from seeded data with summary, metrics table, findings, recommendations, and limitations.",
+    "files": [
+      {
+        "path": "data/returns-summary.csv",
+        "content": "month,region,return_rate,refund_leakage,response_hours,escalations\nMarch,West,0.112,0.142,18,44\nApril,West,0.108,0.101,9,31\nMay,West,0.099,0.089,6,25\nMarch,East,0.087,0.096,15,21\nApril,East,0.091,0.094,13,24\nMay,East,0.088,0.092,12,22\n"
+      }
+    ],
+    "worker": {
+      "name": "Mats",
+      "role": "Data Analyst"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "workspace/report.md is a narrative analytical report with metrics table, findings, recommendations, and limitations grounded in seeded data.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/report.md",
+        "kind": "markdown-report",
+        "minBytes": 2200,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "Executive summary|Key metrics|Findings|Recommendations|Limitations",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "March|April|May|West|East"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "14\\.2%|8\\.9%|18|6|44|25",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "refund leakage|response hours|escalations|return rate",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "table|\\|",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/report.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "accuracy",
+        "description": "Computed figures are correct against the seeded records."
+      },
+      {
+        "name": "grounding",
+        "description": "Every claim traces to the seeded data — nothing invented."
+      },
+      {
+        "name": "structure",
+        "description": "Tables and sections make the analysis checkable at a glance."
+      },
+      {
+        "name": "actionability",
+        "description": "Recommendations follow from the findings and are concretely doable."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "analysis before narrative",
+    "source-grounded figures",
+    "limitations"
+  ]
+}