@bendyline/gilde 0.1.55 → 0.1.56

package/data/craftbook-templates/en/engineering-retrospective/versions/2.0.8/test.json

@@ -0,0 +1,191 @@
+{
+  "schemaVersion": 1,
+  "title": "Engineering Retrospective — queue reliability window",
+  "objective": "Exercise the Engineering Retrospective craftbook on a fixed exported history and require reconciled metrics, contextual interpretation, balanced growth notes, and owned experiments.",
+  "tags": [
+    "workflow",
+    "retrospective",
+    "engineering",
+    "corpus"
+  ],
+  "prompt": "Use the Engineering Retrospective craftbook for the fixed 2026-07-01 through 2026-07-14 window. This evaluation has no live git repository: `source/git-log.txt` and `source/numstat.tsv` are the authoritative read-only exports, and `source/team-events.md` supplies context. Do not treat unavailable live history as zero or ask for another window. Reconcile human activity, exclude the release bot, identify evidence-backed hotspots and quality/collaboration signals, and write the workflow artifacts ending at `retro/engineering-retrospective.md`. Include constructive notes for Alex Kim and Priya Shah without rankings or volume-as-productivity claims, then propose three to five owner-backed experiments with measurable outcomes and a review date.",
+  "setup": {
+    "projectName": "Queue reliability engineering retrospective",
+    "about": "A self-contained retrospective over exported repository evidence. The export and context files are complete; no network, live git repository, or personnel system is available.",
+    "missionObjectives": "Run the Engineering Retrospective craftbook as a real task, reconcile the fixed history, preserve evidence limits, and produce an actionable, fair retrospective.",
+    "files": [
+      {
+        "path": "source/git-log.txt",
+        "content": "# sha|date|author|email|subject\na1c001|2026-07-02|Alex Kim|alex@example.test|fix queue timeout classification\na1c002|2026-07-03|Priya Shah|priya@example.test|add queue retry metrics\na1c003|2026-07-06|Priya Shah|priya@example.test|document on-call queue recovery\na1c004|2026-07-09|Alex Kim|alex@example.test|cover refresh-token expiry\na1c005|2026-07-10|Alex Kim|alex@example.test|add queue timeout regression fixture\na1c006|2026-07-12|Priya Shah|priya@example.test|pin parser dependency after advisory review\na1c007|2026-07-14|release-bot|bot@example.test|release 2.8.0\n"
+      },
+      {
+        "path": "source/numstat.tsv",
+        "content": "sha\tadded\tdeleted\tpath\na1c001\t34\t12\tsrc/queue.ts\na1c001\t22\t0\ttests/queue.test.ts\na1c002\t28\t4\tsrc/queue.ts\na1c002\t40\t2\tsrc/metrics.ts\na1c003\t55\t3\tdocs/oncall.md\na1c004\t18\t7\tsrc/auth.ts\na1c004\t26\t0\ttests/auth.test.ts\na1c005\t12\t6\tsrc/queue.ts\na1c005\t31\t2\ttests/queue.test.ts\na1c006\t3\t3\tpackage.json\na1c006\t14\t14\tpnpm-lock.yaml\na1c007\t2\t2\tCHANGELOG.md\n"
+      },
+      {
+        "path": "source/team-events.md",
+        "content": "# Team context\n\n- Window: 2026-07-01 through 2026-07-14, UTC. Compare only within this window; no prior baseline export is available.\n- Queue retry incident on 2026-07-02 caused delayed job status for 41 minutes. Priya Shah was primary responder; Alex Kim traced the timeout classification the next morning.\n- Alex was on planned leave July 4-5.\n- Priya paired with Alex on the queue regression fixture; pairing is mentioned in commit a1c005 but the export cannot attribute line-level co-authorship.\n- Release-bot is automation and must not be included in human activity or growth notes.\n- Review comments, meetings, and off-repository operational work are not present in this export.\n- Experiments should be reviewed on 2026-07-28.\n"
+      }
+    ]
+  },
+  "mocks": [],
+  "success": {
+    "summary": "The real craftbook reconciles the exported history into a contextual, non-ranking retrospective with balanced growth notes and measurable experiments.",
+    "deliverables": [
+      {
+        "path": "retro/scope.md",
+        "kind": "markdown-notes",
+        "minBytes": 600,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "retro/scope.md",
+            "pattern": "2026-07-01[\\s\\S]*2026-07-14[\\s\\S]*(Alex Kim|alex@example\\.test)[\\s\\S]*(Priya Shah|priya@example\\.test)[\\s\\S]*(release-bot|automation)",
+            "flags": "i",
+            "label": "window, identity map, and bot exclusion"
+          }
+        ]
+      },
+      {
+        "path": "retro/evidence.md",
+        "kind": "markdown-report",
+        "minBytes": 1100,
+        "checks": [
+          {
+            "kind": "valueGrounding",
+            "file": "retro/evidence.md",
+            "facts": [
+              {
+                "id": "human-commits",
+                "label": "six human commits",
+                "required": [
+                  "6\\s+human\\s+commits?|six\\s+human\\s+commits?"
+                ]
+              },
+              {
+                "id": "bot-exclusion",
+                "label": "one excluded bot commit",
+                "required": [
+                  "1\\s+(?:bot|automated)|one\\s+(?:bot|automated)"
+                ]
+              },
+              {
+                "id": "alex-count",
+                "label": "three Alex commits",
+                "required": [
+                  "Alex(?: Kim)?[^\\n]{0,80}3"
+                ]
+              },
+              {
+                "id": "priya-count",
+                "label": "three Priya commits",
+                "required": [
+                  "Priya(?: Shah)?[^\\n]{0,80}3"
+                ]
+              },
+              {
+                "id": "hotspot",
+                "label": "queue file changed in three commits",
+                "required": [
+                  "src/queue\\.ts[^\\n]{0,100}3|3[^\\n]{0,100}src/queue\\.ts"
+                ]
+              }
+            ]
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "retro/evidence.md",
+            "minCitations": 3
+          }
+        ]
+      },
+      {
+        "path": "retro/engineering-retrospective.md",
+        "kind": "markdown-report",
+        "minBytes": 1600,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "retro/engineering-retrospective.md",
+            "pattern": "^#{1,3}\\s+Period context\\b[\\s\\S]*^#{1,3}\\s+What went well\\b[\\s\\S]*^#{1,3}\\s+Friction and hotspots\\b[\\s\\S]*^#{1,3}\\s+Collaboration and growth\\b[\\s\\S]*^#{1,3}\\s+Experiments\\b[\\s\\S]*^#{1,3}\\s+Evidence limits\\b",
+            "flags": "im",
+            "label": "complete retrospective structure"
+          },
+          {
+            "kind": "contains",
+            "file": "retro/engineering-retrospective.md",
+            "pattern": "Alex Kim[\\s\\S]*Priya Shah|Priya Shah[\\s\\S]*Alex Kim",
+            "label": "balanced named growth notes"
+          },
+          {
+            "kind": "contains",
+            "file": "retro/engineering-retrospective.md",
+            "pattern": "2026-07-28[\\s\\S]*(owner|Alex|Priya)|(owner|Alex|Priya)[\\s\\S]*2026-07-28",
+            "flags": "i",
+            "label": "owned experiments with review date"
+          },
+          {
+            "kind": "notContains",
+            "file": "retro/engineering-retrospective.md",
+            "pattern": "top performer|bottom performer|most productive|least productive|performance score|rank(?:ed|ing)?\\s+(?:the\\s+)?(?:engineers|people|team)",
+            "flags": "i",
+            "label": "no personnel ranking or volume-as-productivity framing"
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "retro/engineering-retrospective.md",
+            "minCitations": 3
+          }
+        ]
+      }
+    ],
+    "taskNotes": {
+      "minBytes": 180,
+      "checks": [
+        {
+          "kind": "contains",
+          "file": "task-notes.md",
+          "pattern": "\\bDONE\\b[\\s\\S]*retro/engineering-retrospective\\.md[\\s\\S]*(experiment|evidence)",
+          "flags": "i",
+          "label": "terminal note records the evidence-backed retrospective"
+        }
+      ],
+      "requireCraftbookTask": true
+    },
+    "taskGraph": {
+      "requireCraftbookTask": true,
+      "requireTerminalStep": true
+    },
+    "unchangedFixtures": [
+      "source/git-log.txt",
+      "source/numstat.tsv",
+      "source/team-events.md"
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "retro/engineering-retrospective.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "Evidence integrity",
+        "description": "Counts reconcile to the exports, bot activity is excluded, hotspots are traceable, and missing review or operational data is not treated as zero."
+      },
+      {
+        "name": "Context and fairness",
+        "description": "The incident, leave, pairing, and evidence limits shape interpretation; individual notes are constructive and do not rank people by volume."
+      },
+      {
+        "name": "Actionability",
+        "description": "The retrospective turns patterns into a small set of owner-backed, measurable experiments with a fixed review date."
+      }
+    ],
+    "contextNote": "The three source exports are the complete evidence set. Do not reward invented review metrics, cycle time, or off-repository work."
+  },
+  "qualityFocus": [
+    "Reconcile six human commits and exclude release automation",
+    "Explain the queue hotspot in incident context without equating volume with value",
+    "Give Alex and Priya balanced growth notes and measurable experiments"
+  ]
+}

package/data/craftbook-templates/ep/episode-plan/versions/1.0.3/test.json

@@ -41,30 +41,35 @@
             "kind": "contains",
             "file": "tasks/eval/report.md",
             "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
-            "flags": "i"
+            "flags": "i",
+            "artifact": true
           },
           {
             "kind": "contains",
             "file": "tasks/eval/report.md",
             "pattern": "segment|rundown",
             "flags": "i",
-            "label": "rundown present"
+            "label": "rundown present",
+            "artifact": true
           },
           {
             "kind": "contains",
             "file": "tasks/eval/report.md",
             "pattern": "\\d{1,2}\\s*(min|:)",
             "flags": "i",
-            "label": "minute marks on segments"
+            "label": "minute marks on segments",
+            "artifact": true
           },
           {
             "kind": "contains",
             "file": "tasks/eval/report.md",
             "pattern": "prep|before record",
             "flags": "i",
-            "label": "prep list present"
+            "label": "prep list present",
+            "artifact": true
           }
-        ]
+        ],
+        "artifact": true
       }
     ]
   },

package/data/craftbook-templates/ex/executive-level-review/versions/2.0.7/test.json

@@ -18,38 +18,35 @@
         "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"
       }
-    ],
-    "craftbookParams": {
-      "workPath": "tasks/eval"
-    }
+    ]
   },
   "mocks": [],
   "success": {
     "summary": "The named craftbook produces a grounded executive verdict, alternatives, scorecard, and terminal task record.",
     "deliverables": [
       {
-        "path": "tasks/eval/reviews/executive-level-review.md",
+        "path": "reviews/executive-level-review.md",
         "kind": "markdown-report",
         "artifact": true,
         "minBytes": 1400,
         "checks": [
           {
             "kind": "contains",
-            "file": "tasks/eval/reviews/executive-level-review.md",
+            "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": "tasks/eval/reviews/executive-level-review.md",
+            "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": "tasks/eval/reviews/executive-level-review.md",
+            "file": "reviews/executive-level-review.md",
             "facts": [
               {
                 "id": "customer-requests",
@@ -85,7 +82,7 @@
           },
           {
             "kind": "citationsResolve",
-            "file": "tasks/eval/reviews/executive-level-review.md",
+            "file": "reviews/executive-level-review.md",
             "minCitations": 1
           }
         ]
@@ -113,7 +110,7 @@
   },
   "rubric": {
     "artifact": {
-      "path": "tasks/eval/reviews/executive-level-review.md",
+      "path": "reviews/executive-level-review.md",
       "kind": "markdown"
     },
     "axes": [