@bendyline/gilde 0.1.60 → 0.1.62

package/data/craftbook-templates/po/powerpoint-deck/versions/1.7.11/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"
+  ]
+}