@bendyline/gilde 0.1.57 → 0.1.59

package/data/craftbook-templates/po/powerpoint-deck/versions/1.7.10/test.json

@@ -0,0 +1,240 @@
+{
+  "schemaVersion": 1,
+  "title": "PowerPoint source-acquisition, outline parity, and DocBlocks production",
+  "objective": "Measure the whole PowerPoint route in one eval: acquire and cite sources, hold the deck to its own outline, and then actually PRODUCE the binary — 1.5.0 graded only the Markdown and 1.4.0 graded only the production, so each on its own let half the recipe pass unverified.",
+  "tags": [
+    "corpus",
+    "research",
+    "grounding",
+    "external",
+    "documents"
+  ],
+  "prompt": "Create a PowerPoint from source/brief.md for operations leads. Use 6-8 slides, preserve the supplied facts exactly, and close with next actions. Do not create HTML.\n\nProduce the finished file through the DocBlocks tools on your roster: `list_roots`, then `convert_document` on the approved `powerpoint/eval/deck.md`, `preview_document` to confirm it rendered, and `save_artifact` to write the deck to `deliverables/d-day.pptx`. Markdown or hand-built OOXML written to that path is not a substitute for the real file.",
+  "setup": {
+    "projectName": "PowerPoint Grounding Eval",
+    "about": "The exact source file is authoritative; the deck must keep the locked outline and its facts.",
+    "missionObjectives": "Read source/brief.md, cite it, lock a numbered outline, and create matching slide-structured Markdown ready for editable PPTX conversion.",
+    "craftbookParams": {
+      "outputPath": "deliverables/d-day.pptx",
+      "sourcePath": "source/brief.md",
+      "topic": "Boreal Desk Returns Pilot",
+      "content": "",
+      "audience": "operations leads",
+      "outputDir": "powerpoint/eval"
+    },
+    "files": [
+      {
+        "path": "source/brief.md",
+        "content": "# Boreal Desk Returns Pilot\n\nAudience: operations leads.\n\nFacts:\n- Pilot coverage: 18 SKUs.\n- Median first response: 18 hours to 6 hours.\n- Preventable refund leakage: 14.2% to 8.9%.\n- Top unresolved complaint: status silence after photo submission.\n- Next actions: automated status emails, barcode-exception training, weekly Finance exception export.\n\nDo not invent other metrics."
+      }
+    ],
+    "worker": {
+      "name": "Mara",
+      "role": "Researcher"
+    }
+  },
+  "mocks": [
+    {
+      "kind": "mcp",
+      "id": "docblocks",
+      "toolsetId": "docblocks",
+      "description": "Deterministic fake DocBlocks MCP for final presentation conversion, preview, and saving.",
+      "tools": [
+        {
+          "name": "list_roots",
+          "description": "List the workspace read root and artifacts write root.",
+          "resultTemplate": {
+            "roots": [
+              {
+                "id": "workspace",
+                "name": "workspace",
+                "capabilities": [
+                  "read"
+                ]
+              },
+              {
+                "id": "artifacts",
+                "name": "artifacts",
+                "capabilities": [
+                  "read",
+                  "write"
+                ]
+              }
+            ]
+          }
+        },
+        {
+          "name": "get_authoring_context",
+          "description": "Return optional PPTX authoring guidance.",
+          "resultTemplate": {
+            "targetFormat": "pptx",
+            "guidance": "Use one H1 per slide and editable-native fidelity."
+          }
+        },
+        {
+          "name": "convert_document",
+          "description": "Convert approved Markdown to a deterministic PPTX artifact URI.",
+          "resultTemplate": {
+            "artifacts": [
+              {
+                "format": "pptx",
+                "uri": "mock://docblocks/deck.pptx",
+                "sha256": "1111111111111111111111111111111111111111111111111111111111111111"
+              }
+            ],
+            "diagnostics": []
+          }
+        },
+        {
+          "name": "preview_document",
+          "description": "Return deterministic slide-preview evidence.",
+          "resultTemplate": {
+            "slides": 7,
+            "overflow": false,
+            "tinyText": false,
+            "previewed": true
+          }
+        },
+        {
+          "name": "save_artifact",
+          "description": "Save the converted PPTX into the project artifacts drawer.",
+          "resultTemplate": {
+            "ok": true,
+            "bytes": 2400,
+            "sha256": "2222222222222222222222222222222222222222222222222222222222222222"
+          },
+          "writeFixture": {
+            "surface": "artifact",
+            "pathArgument": "destination.path",
+            "fixture": "minimal-pptx"
+          }
+        }
+      ]
+    }
+  ],
+  "success": {
+    "summary": "The source packet cites the exact brief, the outline is numbered, powerpoint/eval/deck.md matches every outline slide while preserving the supplied facts, and a real ZIP-shaped PPTX — not the Markdown source — lands at the requested workspace path via DocBlocks.",
+    "deliverables": [
+      {
+        "path": "{{task.dir}}/sources.md",
+        "kind": "markdown-doc",
+        "minBytes": 500,
+        "artifact": true,
+        "checks": [
+          {
+            "kind": "citationsResolve",
+            "file": "{{task.dir}}/sources.md",
+            "artifact": true,
+            "minCitations": 1
+          }
+        ]
+      },
+      {
+        "path": "{{task.dir}}/outline.md",
+        "kind": "markdown-doc",
+        "minBytes": 500,
+        "artifact": true,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "{{task.dir}}/outline.md",
+            "pattern": "(?:^|\\n)#{2,6}\\s+(?:Slide\\s+)?1(?:\\s*[.:\\-–—]\\s*|\\s+)\\S",
+            "flags": "i",
+            "label": "a numbered Slide 1 heading",
+            "artifact": true
+          }
+        ]
+      },
+      {
+        "path": "powerpoint/eval/deck.md",
+        "kind": "markdown-doc",
+        "minBytes": 700,
+        "checks": [
+          {
+            "kind": "markdownHeadingsMatch",
+            "file": "powerpoint/eval/deck.md",
+            "outlineFile": "{{task.dir}}/outline.md",
+            "outlineArtifact": true
+          },
+          {
+            "kind": "citationsResolve",
+            "file": "powerpoint/eval/deck.md",
+            "minCitations": 1
+          },
+          {
+            "kind": "contains",
+            "file": "powerpoint/eval/deck.md",
+            "pattern": "18\\s+SKUs|18\\s+hours|6\\s+hours|14\\.2%|8\\.9%",
+            "flags": "i",
+            "label": "the supplied figures, verbatim"
+          },
+          {
+            "kind": "contains",
+            "file": "powerpoint/eval/deck.md",
+            "pattern": "automated status emails|barcode-exception training|weekly Finance exception export",
+            "flags": "i",
+            "label": "the supplied next actions"
+          },
+          {
+            "kind": "notContains",
+            "file": "powerpoint/eval/deck.md",
+            "pattern": "<html|<script|<style",
+            "flags": "i"
+          }
+        ]
+      },
+      {
+        "path": "deliverables/d-day.pptx",
+        "kind": "slide-deck",
+        "minBytes": 1000,
+        "checks": [
+          {
+            "kind": "binaryDocument",
+            "file": "deliverables/d-day.pptx",
+            "minBytes": 1000
+          }
+        ]
+      }
+    ],
+    "mocks": [
+      {
+        "service": "docblocks",
+        "requiredTools": [
+          "convert_document",
+          "preview_document",
+          "save_artifact"
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "powerpoint/eval/deck.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "source-acquisition",
+        "description": "The workflow read and cited the exact supplied source before outlining."
+      },
+      {
+        "name": "outline-parity",
+        "description": "Every numbered outline slide appears exactly once as a same-title H1 in powerpoint/eval/deck.md, in order."
+      },
+      {
+        "name": "grounding",
+        "description": "Every metric and action comes from source/brief.md and no extra metric was invented."
+      },
+      {
+        "name": "convertibility",
+        "description": "One H1 per slide and concise block content make the source ready for DocBlocks PPTX conversion."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "successful exact-source retrieval",
+    "resolvable citations",
+    "exact outline-to-deck slide parity",
+    "no HTML intermediary"
+  ]
+}

package/data/craftbook-templates/pr/practice-exam/versions/1.0.3/test.json

@@ -19,14 +19,11 @@
     "worker": {
       "name": "Milan",
       "role": "Developer"
-    },
-    "craftbookParams": {
-      "workPath": "tasks/eval"
     }
   },
   "mocks": [],
   "success": {
-    "summary": "exam.html is a timed, self-scoring exam built from the deck with the record's weak topics weighted up; tasks/eval/answer-key.md explains every answer.",
+    "summary": "exam.html is a timed, self-scoring exam built from the deck with the record's weak topics weighted up; answer-key.md in this task's own artifacts working folder explains every answer.",
     "deliverables": [
       {
         "path": "exam.html",
@@ -73,13 +70,13 @@
         ]
       },
       {
-        "path": "tasks/eval/answer-key.md",
+        "path": "{{task.dir}}/answer-key.md",
         "kind": "markdown-notes",
         "minBytes": 800,
         "checks": [
           {
             "kind": "contains",
-            "file": "tasks/eval/answer-key.md",
+            "file": "{{task.dir}}/answer-key.md",
             "pattern": "card|deck",
             "flags": "i",
             "label": "key points back to the deck",

package/data/craftbook-templates/pr/practice-session/versions/1.0.3/test.json

@@ -23,30 +23,27 @@
     "worker": {
       "name": "Ruben",
       "role": "Coach"
-    },
-    "craftbookParams": {
-      "workPath": "tasks/eval"
     }
   },
   "mocks": [],
   "success": {
-    "summary": "Today's session is appended to practice.json as a well-formed entry, and tasks/eval/report.md closes the day with the streak state and a concrete, targeted plan for tomorrow.",
+    "summary": "Today's session is appended to practice.json as a well-formed entry, and report.md in this task's own artifacts working folder closes the day with the streak state and a concrete, targeted plan for tomorrow.",
     "deliverables": [
       {
-        "path": "tasks/eval/report.md",
+        "path": "{{task.dir}}/report.md",
         "kind": "markdown-report",
         "minBytes": 800,
         "checks": [
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
             "flags": "i",
             "artifact": true
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "tomorrow|next session",
             "flags": "i",
             "label": "tomorrow's plan present",
@@ -54,7 +51,7 @@
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "bpm|bar\\s*\\d+",
             "flags": "i",
             "label": "concrete tempo or bar targets",
@@ -91,7 +88,7 @@
   },
   "rubric": {
     "artifact": {
-      "path": "tasks/eval/report.md",
+      "path": "{{task.dir}}/report.md",
       "kind": "markdown"
     },
     "axes": [

package/data/craftbook-templates/pr/price-tracker/versions/1.0.3/craftbook.json

@@ -104,14 +104,6 @@
             "kind": "sniff",
             "file": "price_tracker.py",
             "sniff": "nonempty"
-          },
-          {
-            "kind": "esmImports",
-            "file": "price_tracker.py"
-          },
-          {
-            "kind": "sourceParses",
-            "file": "price_tracker.py"
           }
         ],
         "onReject": "build",

package/data/craftbook-templates/re/receipt-intake/versions/1.0.3/test.json

@@ -8,8 +8,8 @@
   "prompt": "i dropped this months recipts into inbox/ — can you get them into the shoebox and flag anything that looks off?",
   "setup": {
     "projectName": "Tax Shoebox — Receipt Intake Eval",
-    "about": "Self-contained eval project for receipt-intake. The intake rail has already pre-extracted each snapped receipt photo into a plain-text note under inbox/ (vendor, date, total, and any handwriting). The ledger lives at ledger.json in the workspace; the run's write-up goes to tasks/eval/report.md.",
-    "missionObjectives": "Use the Receipt Intake craftbook/template, read the seeded local fixtures only, and finish with an updated ledger.json plus tasks/eval/report.md — no network calls, real credentials, or live services.",
+    "about": "Self-contained eval project for receipt-intake. The intake rail has already pre-extracted each snapped receipt photo into a plain-text note under inbox/ (vendor, date, total, and any handwriting). The ledger lives at ledger.json in the workspace; the run's write-up goes to report.md in this task's own artifacts working folder.",
+    "missionObjectives": "Use the Receipt Intake craftbook/template, read the seeded local fixtures only, and finish with an updated ledger.json plus report.md in this task's own artifacts working folder — no network calls, real credentials, or live services.",
     "files": [
       {
         "path": "inbox/receipt-01.txt",
@@ -39,30 +39,27 @@
     "worker": {
       "name": "Nora",
       "role": "Clerk"
-    },
-    "craftbookParams": {
-      "workPath": "tasks/eval"
     }
   },
   "mocks": [],
   "success": {
-    "summary": "The ledger gains schema-clean entries whose amounts match the receipts exactly, and tasks/eval/report.md surfaces the duplicate and the owner question instead of guessing.",
+    "summary": "The ledger gains schema-clean entries whose amounts match the receipts exactly, and report.md in this task's own artifacts working folder surfaces the duplicate and the owner question instead of guessing.",
     "deliverables": [
       {
-        "path": "tasks/eval/report.md",
+        "path": "{{task.dir}}/report.md",
         "kind": "markdown-report",
         "minBytes": 600,
         "checks": [
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
             "flags": "i",
             "artifact": true
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "question|unsure|unclear|flag",
             "flags": "i",
             "label": "ambiguous receipt surfaced as an owner question, not a guess",
@@ -70,7 +67,7 @@
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "duplicate|already (in|entered|logged)|double[- ]?count",
             "flags": "i",
             "label": "the repeat Office Depot receipt is flagged as a likely duplicate",
@@ -126,7 +123,7 @@
   },
   "rubric": {
     "artifact": {
-      "path": "tasks/eval/report.md",
+      "path": "{{task.dir}}/report.md",
       "kind": "markdown"
     },
     "axes": [

package/data/craftbook-templates/re/recipe-capture/versions/1.0.3/test.json

@@ -8,8 +8,8 @@
   "prompt": "scanned two of omas cards — the transcriptions are in inbox/card-1.txt and inbox/card-2.txt. can you get them into the book propperly?",
   "setup": {
     "projectName": "Family Recipe Book — Card Capture Eval",
-    "about": "Self-contained eval project for recipe-capture. The card photos have already been pre-extracted into plain-text scan notes under inbox/ (the text as written, plus margin notes). Captured recipes go under recipes/ and are recorded in recipes.json; the capture log goes to tasks/eval/report.md.",
-    "missionObjectives": "Use the Capture a Recipe Card craftbook/template, work only from the seeded card notes, and finish with the recipe files, an updated recipes.json, and tasks/eval/report.md — no network calls, real credentials, or live services.",
+    "about": "Self-contained eval project for recipe-capture. The card photos have already been pre-extracted into plain-text scan notes under inbox/ (the text as written, plus margin notes). Captured recipes go under recipes/ and are recorded in recipes.json; the capture log goes to report.md in this task's own artifacts working folder.",
+    "missionObjectives": "Use the Capture a Recipe Card craftbook/template, work only from the seeded card notes, and finish with the recipe files, an updated recipes.json, and report.md in this task's own artifacts working folder — no network calls, real credentials, or live services.",
     "files": [
       {
         "path": "inbox/card-1.txt",
@@ -27,9 +27,6 @@
     "worker": {
       "name": "Bep",
       "role": "Archivist"
-    },
-    "craftbookParams": {
-      "workPath": "tasks/eval"
     }
   },
   "mocks": [],
@@ -37,34 +34,34 @@
     "summary": "Both cards become recipe files under recipes/ with provenance kept beside the food, the store records each capture, and the log flags what needs a test-cook.",
     "deliverables": [
       {
-        "path": "tasks/eval/report.md",
+        "path": "{{task.dir}}/report.md",
         "kind": "markdown-report",
         "minBytes": 500,
         "checks": [
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
             "flags": "i",
             "artifact": true
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "Oma|Riet",
             "label": "provenance carried into the capture log",
             "artifact": true
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "\\?",
             "label": "questions for the family are actually asked",
             "artifact": true
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "test[- ]?cook|try (it|the)|verify",
             "flags": "i",
             "label": "test-cook needs named",
@@ -106,7 +103,7 @@
   },
   "rubric": {
     "artifact": {
-      "path": "tasks/eval/report.md",
+      "path": "{{task.dir}}/report.md",
       "kind": "markdown"
     },
     "axes": [

package/data/craftbook-templates/re/record-a-relative/versions/1.0.3/test.json

@@ -23,29 +23,26 @@
     "worker": {
       "name": "Teun",
       "role": "Genealogist"
-    },
-    "craftbookParams": {
-      "workPath": "tasks/eval"
     }
   },
   "mocks": [],
   "success": {
-    "summary": "The transcript is filed under interviews/, and tasks/eval/report.md extracts the facts with their quotes, surfaces the 1919-vs-1921 birth-year conflict as a conflict, keeps the uncertain brother story uncertain, and asks follow-up questions.",
+    "summary": "The transcript is filed under interviews/, and report.md in this task's own artifacts working folder extracts the facts with their quotes, surfaces the 1919-vs-1921 birth-year conflict as a conflict, keeps the uncertain brother story uncertain, and asks follow-up questions.",
     "deliverables": [
       {
-        "path": "tasks/eval/report.md",
+        "path": "{{task.dir}}/report.md",
         "kind": "markdown-report",
         "minBytes": 1000,
         "checks": [
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
             "flags": "i"
           },
           {
             "kind": "valuesSubsetOf",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "sourceFiles": [
               "recordings/oma-2026-07.txt",
               "tree.json"
@@ -56,18 +53,19 @@
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "conflict|differs|doesn.t match|discrepan|disagree",
             "flags": "i",
             "label": "the tree conflict is surfaced"
           },
           {
             "kind": "contains",
-            "file": "tasks/eval/report.md",
+            "file": "{{task.dir}}/report.md",
             "pattern": "\\?",
             "label": "follow-up questions are real questions"
           }
-        ]
+        ],
+        "artifact": true
       }
     ],
     "checks": [
@@ -83,7 +81,7 @@
   },
   "rubric": {
     "artifact": {
-      "path": "tasks/eval/report.md",
+      "path": "{{task.dir}}/report.md",
       "kind": "markdown"
     },
     "axes": [

package/data/craftbook-templates/re/recurring-invoice-run/versions/1.0.4/craftbook.json

@@ -129,14 +129,6 @@
             "kind": "sniff",
             "file": "invoice_run.py",
             "sniff": "nonempty"
-          },
-          {
-            "kind": "esmImports",
-            "file": "invoice_run.py"
-          },
-          {
-            "kind": "sourceParses",
-            "file": "invoice_run.py"
           }
         ],
         "onReject": "build",
@@ -194,14 +186,6 @@
             "kind": "sniff",
             "file": "invoice_run.py",
             "sniff": "nonempty"
-          },
-          {
-            "kind": "esmImports",
-            "file": "invoice_run.py"
-          },
-          {
-            "kind": "sourceParses",
-            "file": "invoice_run.py"
           }
         ],
         "onReject": "verify",