@bendyline/gilde 0.1.26 → 0.1.28

package/data/craftbook-templates/ex/executive-level-review/versions/2.0.3/test.json

@@ -0,0 +1,136 @@
+{
+  "schemaVersion": 1,
+  "title": "Executive-Level Review — analytics expansion decision",
+  "objective": "Test whether the Executive-Level Review craftbook challenges a proposal, compares lower-cost alternatives, and produces an evidence-anchored mode verdict with decision conditions.",
+  "tags": [
+    "workflow",
+    "strategy",
+    "executive-review",
+    "decision"
+  ],
+  "prompt": "Use the Executive-Level Review craftbook to review `source/analytics-proposal.md`. Challenge whether the proposed scope is the best way to reach the outcome; compare at least three alternatives including the two-day manual export and waiting. Produce all craftbook artifacts and a decision-ready `reviews/executive-level-review.md`. Use exactly one allowed mode verdict, score every required dimension from 0–10 with a factual basis, separate unknowns from evidence, and state explicit approve, pause, and stop conditions. Do not invent customer or financial evidence.",
+  "setup": {
+    "projectName": "Analytics expansion review",
+    "about": "A hermetic executive decision review. The seeded proposal is the complete evidence set.",
+    "missionObjectives": "Decide what, if anything, should be approved now while preserving the fastest credible route to learning.",
+    "files": [
+      {
+        "path": "source/analytics-proposal.md",
+        "content": "# Embedded analytics proposal\n\n## Requested decision\nApprove an eight-week build of customer-facing dashboards with scheduled PDF delivery and CSV export.\n\n## Evidence\nTwelve of 86 active customers requested better reporting in support conversations. Three customers agreed to a design pilot, but none signed a paid commitment. Support currently creates weekly exports for five customers.\n\n## Cost and constraints\n- Estimate: two engineers for 8 weeks.\n- New charting service: $4,000 per month at current volume.\n- Security review is complete; accessibility and data-retention reviews are not.\n- The enterprise renewal decision is in six weeks.\n- A manual scheduled-export workflow can be built in two engineer-days and would test delivery cadence, but not dashboard engagement.\n\n## Unknowns\nWillingness to pay, which metrics matter, PDF accessibility, and whether scheduled delivery or interactive dashboards drives retention.\n"
+      }
+    ]
+  },
+  "mocks": [],
+  "success": {
+    "summary": "The named craftbook produces a grounded executive verdict, alternatives, scorecard, and terminal task record.",
+    "deliverables": [
+      {
+        "path": "reviews/executive-level-review.md",
+        "kind": "markdown-report",
+        "artifact": true,
+        "minBytes": 1400,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "reviews/executive-level-review.md",
+            "pattern": "^#{1,3}\\s+Executive verdict\\b[\\s\\S]*(SCOPE EXPANSION|SELECTIVE|HOLD|SCOPE REDUCTION)[\\s\\S]*^#{1,3}\\s+Scorecard\\b[\\s\\S]*^#{1,3}\\s+Alternatives\\b[\\s\\S]*^#{1,3}\\s+Recommendation\\b[\\s\\S]*^#{1,3}\\s+Decision conditions\\b",
+            "flags": "im",
+            "label": "verdict, scorecard, alternatives, recommendation, and conditions"
+          },
+          {
+            "kind": "contains",
+            "file": "reviews/executive-level-review.md",
+            "pattern": "(user value).*?(strategic fit).*?(evidence strength).*?(feasibility).*?(sequencing).*?(reversibility).*?(operating cost).*?(downside)",
+            "flags": "is",
+            "label": "complete executive scorecard"
+          },
+          {
+            "kind": "valueGrounding",
+            "file": "reviews/executive-level-review.md",
+            "facts": [
+              {
+                "id": "customer-requests",
+                "required": [
+                  "12\\s+(?:of\\s+86\\s+)?(?:active\\s+)?customers"
+                ]
+              },
+              {
+                "id": "pilot-evidence",
+                "required": [
+                  "3\\s+customers?[^.]{0,60}(pilot|agreed)"
+                ]
+              },
+              {
+                "id": "build-duration",
+                "required": [
+                  "8\\s+weeks?"
+                ]
+              },
+              {
+                "id": "operating-cost",
+                "required": [
+                  "\\$4,?000\\s+per\\s+month"
+                ]
+              },
+              {
+                "id": "cheap-alternative",
+                "required": [
+                  "(two|2)[ -]?(engineer[- ])?days?[\\s\\S]{0,120}(manual|scheduled)[ -]export"
+                ]
+              }
+            ]
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "reviews/executive-level-review.md",
+            "minCitations": 1
+          }
+        ]
+      }
+    ],
+    "taskNotes": {
+      "minBytes": 160,
+      "checks": [
+        {
+          "kind": "contains",
+          "file": "task-notes.md",
+          "pattern": "\\bDONE\\b[\\s\\S]*reviews/executive-level-review\\.md",
+          "label": "terminal craftbook note names the review"
+        }
+      ],
+      "requireCraftbookTask": true
+    },
+    "taskGraph": {
+      "requireCraftbookTask": true,
+      "requireTerminalStep": true
+    },
+    "unchangedFixtures": [
+      "source/analytics-proposal.md"
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "reviews/executive-level-review.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "Premise challenge",
+        "description": "The review tests whether action is warranted now and compares materially different ways to learn or deliver value."
+      },
+      {
+        "name": "Evidence discipline",
+        "description": "Scores and conclusions trace to supplied facts, with unsupported assumptions and missing decisions called out."
+      },
+      {
+        "name": "Executive usefulness",
+        "description": "The verdict is decisive, sequenced, reversible where possible, and paired with clear decision-changing conditions."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "One allowed mode verdict is selected",
+    "All scorecard dimensions have evidence-backed scores",
+    "The two-day alternative and missing willingness-to-pay evidence influence the decision"
+  ]
+}

package/data/craftbook-templates/ex/expense-categorize/versions/1.0.1/craftbook.json

@@ -0,0 +1,140 @@
+{
+  "id": "expense-categorize",
+  "name": "Categorize Transactions",
+  "description": "Classify a list of bank/card transactions into a fixed chart-of-accounts category set, with confidence and a review queue for the uncertain ones, producing a clean categorized ledger ready for bookkeeping or tax. Locks the category taxonomy and the matching rules (merchant patterns, amount/sign heuristics) FIRST, then categorizes every transaction deterministically-first then by inference, then reviews that the taxonomy is closed, low-confidence rows are queued not force-fit, and totals reconcile to the input. Rules-before-categorize is the wisdom: a defined taxonomy plus rule precedence makes results auditable and repeatable, not a one-off guess.\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 rules (planner) — category taxonomy + matching rules + precedence → gated on artifact `notes/rules-scope.md` (markdown-notes)\n2. Categorize transactions (developer) — apply rules, assign category + confidence, queue doubts → gated on `ledger.json` (json)\n3. Review the ledger (reviewer) — closed taxonomy, reconciliation, review queue → gated on artifact `notes/review.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": "rules-scope",
+  "triggers": [
+    "categorize my transactions",
+    "sort expenses by category",
+    "clean up bank transactions",
+    "bookkeeping categorization",
+    "tag spending"
+  ],
+  "steps": [
+    {
+      "id": "rules-scope",
+      "name": "Scope the rules",
+      "description": "category taxonomy + matching rules + precedence",
+      "prompt": "Define the taxonomy and rules before categorizing. Step 1: load the transactions (date, description/merchant, amount, sign) and list the columns. Step 2: define the closed category set (the chart of accounts, e.g. Income, Office, Software, Travel, Meals, Utilities, Transfers, Uncategorized) — every transaction must land in exactly one. Step 3: define matching rules with precedence: explicit merchant→category patterns first, then heuristics (sign for income vs expense, amount bands, recurring-amount detection), and an Uncategorized fallback. Step 4: define the confidence rule and the review threshold (below which a row goes to a review queue rather than being force-categorized). Step 5: write an acceptance-criteria checklist ('every transaction categorized into exactly one in-set category', 'rule precedence applied deterministically', 'low-confidence rows queued for review', 'per-category totals reconcile to the input sum', 'no category invented outside the taxonomy'). Call write_task_note with taxonomy + rules + checklist and write notes/rules-scope.md. No categorizing yet.\n\nThe deliverable `notes/rules-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": "notes/rules-scope.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "notes/rules-scope.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "notes/rules-scope.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "rules-scope",
+        "maxAttempts": 3
+      },
+      "next": "categorize"
+    },
+    {
+      "id": "categorize",
+      "name": "Categorize transactions",
+      "description": "apply rules, assign category + confidence, queue doubts",
+      "prompt": "Categorize every transaction to the locked taxonomy. Step 1: for each transaction, apply the rules in precedence order — explicit merchant match first, then heuristics, then Uncategorized. Step 2: assign exactly one in-set category plus a confidence and a short reason (which rule fired). Step 3: route rows below the review threshold to a review queue flag rather than forcing a category. Step 4: produce the categorized ledger (the input columns plus category, confidence, reason, needs_review) in the chosen format. Step 5: compute per-category totals and a grand total and confirm the grand total equals the sum of the input amounts. Call write_task_note with the output path, the per-category totals, and the count of rows queued for review.",
+      "suggestedRole": "developer",
+      "advanceWhen": {
+        "file": "ledger.json",
+        "minBytes": 1,
+        "sniff": "json-valid"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "ledger.json",
+            "bytes": 2
+          }
+        ],
+        "scripts": [
+          {
+            "name": "checkJsonValid",
+            "scope": "standard",
+            "inputs": {
+              "file": "ledger.json"
+            }
+          }
+        ],
+        "onReject": "categorize",
+        "maxAttempts": 3
+      },
+      "next": "review"
+    },
+    {
+      "id": "review",
+      "name": "Review the ledger",
+      "description": "closed taxonomy, reconciliation, review queue",
+      "prompt": "Review the categorized ledger against the rules scope. Step 1: confirm the file parses and every transaction has exactly one category drawn only from the closed taxonomy — no invented categories. Step 2: confirm rule precedence was applied — spot-check that explicit merchant matches beat heuristics. Step 3: confirm low-confidence rows are flagged needs_review rather than force-fit, and that the queue is non-empty if genuinely ambiguous rows exist. Step 4: confirm reconciliation — per-category totals sum to the grand total and the grand total equals the input sum (flag any penny drift). Step 5: spot-check 5 categorizations for correctness. Write PASS/FAIL per criterion to notes/review.md and task notes; on failure, name the rows and loop back to categorize.\n\nThe deliverable `notes/review.md` lands in the project's artifacts drawer — write it with `write_artifact` and read it back with `read_artifact`; the shipped workspace stays untouched.",
+      "suggestedRole": "reviewer",
+      "advanceWhen": {
+        "file": "notes/review.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "ledger.json",
+            "bytes": 2
+          }
+        ],
+        "scripts": [
+          {
+            "name": "checkJsonValid",
+            "scope": "standard",
+            "inputs": {
+              "file": "ledger.json"
+            }
+          }
+        ],
+        "onReject": "review",
+        "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 ledger.json and verify every criterion from notes/rules-scope.md. Confirm it parses, every transaction has exactly one category from the closed taxonomy (none invented), rule precedence held (explicit merchant matches beat heuristics — spot-check), low-confidence rows are flagged for review rather than force-fit, and totals reconcile (per-category sums equal the grand total which equals the input sum). Write PASS/FAIL per criterion; on any failure, name the rows and loop back to categorize.\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: \"categorize\" })` 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": "ledger.json"
+        }
+      ],
+      "next": "categorize"
+    },
+    {
+      "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": "1.0.1",
+  "releasedAt": "2026-08-14T00:00:00Z"
+}

package/data/craftbook-templates/ex/expense-categorize/versions/1.0.1/test.json

@@ -0,0 +1,104 @@
+{
+  "schemaVersion": 1,
+  "title": "Categorize Transactions smoke eval",
+  "objective": "Self-contained smoke eval for the Categorize Transactions craftbook using the data generic harness.",
+  "tags": [
+    "data"
+  ],
+  "prompt": "I dropped our numbers in source/records.csv. Can you look at the data, analyze it, and put togehter some results in analysis.md?",
+  "setup": {
+    "projectName": "Categorize Transactions Eval",
+    "about": "Self-contained eval project for expense-categorize. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/analysis.md.",
+    "missionObjectives": "Use the Categorize Transactions craftbook/template, read the seeded local fixtures, and write analysis.md without network calls, real credentials, or live services.",
+    "files": [
+      {
+        "path": "source/brief.md",
+        "content": "# Categorize Transactions 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: expense-categorize - Categorize Transactions.\n"
+      },
+      {
+        "path": "source/records.csv",
+        "content": "customer,region,orders,revenue_usd,churn_risk\nAster,North,12,1440,low\nBoreal,South,8,720,medium\nCedar,West,15,2100,high\nDune,North,5,500,low\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Data Analyst"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "analysis.md is a seeded-data analysis with deterministic computed facts.",
+    "deliverables": [
+      {
+        "path": "analysis.md",
+        "kind": "markdown-report",
+        "minBytes": 900,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "analysis.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "analysis.md",
+            "pattern": "40\\b",
+            "flags": "i",
+            "label": "include total orders of 40"
+          },
+          {
+            "kind": "contains",
+            "file": "analysis.md",
+            "pattern": "\\$?4,?760\\b",
+            "flags": "i",
+            "label": "include total revenue of 4760"
+          },
+          {
+            "kind": "contains",
+            "file": "analysis.md",
+            "pattern": "Cedar",
+            "flags": "i",
+            "label": "identify Cedar as top revenue and high risk"
+          },
+          {
+            "kind": "contains",
+            "file": "analysis.md",
+            "pattern": "high",
+            "flags": "i",
+            "label": "include churn-risk finding"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "analysis.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": [
+    "seeded data reading",
+    "computed facts",
+    "recommendation grounding"
+  ]
+}

package/data/craftbook-templates/fa/fact-check/versions/1.0.1/craftbook.json

@@ -0,0 +1,146 @@
+{
+  "id": "fact-check",
+  "name": "Fact-Check a Set of Claims",
+  "description": "Verify a set of claims one by one and produce a verdict report: a planner extracts and itemizes every checkable claim and locks the rating scale, a researcher independently verifies each claim against primary sources and logs the evidence, and a reviewer compiles a per-claim verdict report with ratings and corrections. The ordering matters because verifying before itemizing lets claims slip through unchecked, and a report without a fixed rating scale gives inconsistent verdicts; isolating each atomic claim and locking the scale first makes the check exhaustive and auditable. Use this for fact-checking articles or statements, verifying claims, debunking, accuracy passes, and 'is this true' / 'check these facts' requests.\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. Claim scope (planner) — itemize every checkable claim + lock the rating scale → gated on artifact `notes/claim-scope.md` (markdown-notes)\n2. Verify each claim (researcher) — independently check each claim against primary sources → gated on artifact `notes/evidence.md` (markdown-notes)\n3. Compile the verdict report (reviewer) — per-claim ratings, evidence, and corrections → gated on artifact `factcheck.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": "claim-scope",
+  "triggers": [
+    "fact-check this",
+    "verify these claims",
+    "is this true",
+    "check these facts",
+    "debunk or confirm"
+  ],
+  "steps": [
+    {
+      "id": "claim-scope",
+      "name": "Claim scope",
+      "description": "itemize every checkable claim + lock the rating scale",
+      "prompt": "1. Read the source material and extract EVERY discrete, checkable factual claim into notes/claim-scope.md, each as one atomic statement with an id (C1, C2, ...) — split compound sentences into separate claims and quote the original wording. 2. Mark each claim's type (statistic, attribution, event, causal, definitional) since that drives how it is verified. 3. Lock the rating scale exactly (e.g. True / Mostly True / Mixed / Mostly False / False / Unverifiable) and define each rating. 4. Write an acceptance-criteria checklist (e.g. 'every claim from the source is itemized with an id', 'each claim gets exactly one rating from the locked scale', 'each rating cites at least one independent source', 'false/mixed claims include a correction'). 5. Record decisions. Call write_task_note with the claim list + scale + checklist; write the same to notes/claim-scope.md.\n\nThe deliverable `notes/claim-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": "notes/claim-scope.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "notes/claim-scope.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "notes/claim-scope.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "claim-scope",
+        "maxAttempts": 3
+      },
+      "next": "verify"
+    },
+    {
+      "id": "verify",
+      "name": "Verify each claim",
+      "description": "independently check each claim against primary sources",
+      "prompt": "1. For each claim id, independently verify it against authoritative, primary sources — do NOT trust the source being checked. 2. Log to notes/evidence.md per claim: the id, the source(s) [n] you checked, the exact figure/quote/fact found, whether it confirms, contradicts, or partially supports the claim, and a proposed rating from the locked scale. 3. For numeric or attribution claims, find the original record (the dataset, the filing, the transcript), not a secondary report. 4. If a claim cannot be verified after a genuine search, rate it Unverifiable and say what evidence would settle it. 5. Where a claim is false or mixed, capture the correct fact so the report can state the correction. Good looks like an evidence log where each claim's rating is already justified. Do not write the final report yet.\n\nThe deliverable `notes/evidence.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": "researcher",
+      "advanceWhen": {
+        "file": "notes/evidence.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "notes/evidence.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "notes/evidence.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "verify",
+        "maxAttempts": 3
+      },
+      "next": "report"
+    },
+    {
+      "id": "report",
+      "name": "Compile the verdict report",
+      "description": "per-claim ratings, evidence, and corrections",
+      "prompt": "1. Write factcheck.md with a short summary (counts per rating) then one entry per claim id, in order: the quoted claim, its rating from the locked scale, a one-to-three-sentence justification, the supporting source(s) [n], and — for any False/Mixed claim — the correct statement. 2. Use ONLY ratings from the locked scale and base each strictly on notes/evidence.md; do not introduce new claims or change wordings. 3. Add a References list so every [n] resolves to a real, independent source. 4. Keep verdicts dispassionate and evidence-first. 5. On a loop-back, fix only the named gaps. 6. Call write_task_note with the report path and which criteria now pass.\n\nBefore working, open the artifact input `notes/evidence.md` with `read_artifact`; do not look for it in the workspace.\n\nThe deliverable `factcheck.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": "notes/evidence.md",
+          "artifact": true
+        }
+      ],
+      "advanceWhen": {
+        "file": "factcheck.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "factcheck.md",
+            "bytes": 1500,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "factcheck.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "report",
+        "maxAttempts": 4
+      },
+      "next": "evaluate"
+    },
+    {
+      "id": "evaluate",
+      "name": "Evaluate",
+      "description": "Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.",
+      "prompt": "Open factcheck.md and check EACH criterion PASS/FAIL: (1) every claim id from notes/claim-scope.md has an entry — none dropped; (2) each entry has exactly one rating drawn from the locked scale; (3) each rating cites at least one independent source that is NOT the material being checked; (4) every False/Mixed claim includes a correction; (5) spot-check 3 verdicts against notes/evidence.md to confirm the rating matches the logged evidence; (6) the summary counts match the entries. A report that rates a claim against the source it came from, or skips a claim, FAILS.\n\nThe deliverable lives in the project's artifacts drawer — open `factcheck.md` with `read_artifact`, not `read_file`.\n\nThen route — this is the whole point of the loop:\n\n- **Every criterion PASSES →** call `advance_task_step({ ref, stepId: \"evaluate\", next: \"finish\" })`.\n- **Any criterion FAILS →** write the specific gaps to notes, then call `advance_task_step({ ref, stepId: \"evaluate\", next: \"report\" })` to loop back. The builder fixes exactly those gaps.\n\nNever route to `finish` while any criterion is unmet. The build phase's completion gate already blocked a grossly-incomplete deliverable; your job is the judgment an automated check cannot make (does it actually work, read well, look right). After ~3 unproductive loops, stop and report DONE_WITH_CONCERNS so the user can step in.",
+      "suggestedRole": "reviewer",
+      "consumes": [
+        {
+          "file": "factcheck.md",
+          "artifact": true
+        }
+      ],
+      "next": "report"
+    },
+    {
+      "id": "finish",
+      "name": "Finish",
+      "description": "All acceptance criteria met. Stamp a short summary and report DONE.",
+      "prompt": "Every acceptance criterion passed. Write a one-paragraph DONE summary to task notes via `write_task_note`: what was built, the deliverable path(s), and a one-line confirmation that each criterion is met. Then report DONE.",
+      "suggestedRole": "developer",
+      "terminal": true
+    }
+  ],
+  "version": "1.0.1",
+  "releasedAt": "2026-08-14T00:00:00Z"
+}

package/data/craftbook-templates/fa/fact-check/versions/1.0.1/test.json

@@ -0,0 +1,110 @@
+{
+  "schemaVersion": 1,
+  "title": "Fact-Check a Set of Claims smoke eval",
+  "objective": "Self-contained smoke eval for the Fact-Check a Set of Claims craftbook using the corpus generic harness.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "We need a fact-check a set of claims 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": "Fact-Check a Set of Claims Eval",
+    "about": "Self-contained eval project for fact-check. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/report.md.",
+    "missionObjectives": "Use the Fact-Check a Set of Claims 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": "# Fact-Check a Set of Claims 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: fact-check - Fact-Check a Set of Claims.\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Writer"
+    }
+  },
+  "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"
+  ]
+}