@bendyline/gilde 0.1.50 → 0.1.52

package/data/craftbook-templates/co/competitive-analysis/versions/1.0.3/craftbook.json

@@ -0,0 +1,273 @@
+{
+  "id": "competitive-analysis",
+  "name": "Competitive Analysis",
+  "description": "Compare a set of competitors across consistent dimensions and turn the comparison into an actionable analysis: a planner locks the competitor set and the evaluation dimensions, a researcher fills a feature/positioning matrix with evidence-backed cells, and a copywriter writes the analysis with a clear recommendation and competitive gaps. The ordering matters because dimensions chosen after the research are cherry-picked, and a write-up without a complete matrix has holes a reader will spot; locking the same dimensions for every competitor up front forces an apples-to-apples comparison. Use this for competitor teardowns, market positioning, win/loss analysis, 'how do we compare to X and Y' questions, and feature/pricing comparisons across rivals.\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. Dimension scope (planner) — lock the competitor set + comparison dimensions → gated on artifact `{{workPath}}/dimension-scope.md` (markdown-notes)\n2. Fill the matrix (researcher) — evidence-backed cell for every competitor x dimension → gated on artifact `{{workPath}}/matrix.md` (markdown-notes)\n3. Write the analysis (copywriter) — narrative analysis + recommendation from the matrix → gated on artifact `{{workPath}}/analysis.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": "dimension-scope",
+  "triggers": [
+    "competitive analysis",
+    "compare competitors",
+    "how do we stack up against",
+    "competitor teardown",
+    "market positioning analysis"
+  ],
+  "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}}"
+      }
+    }
+  },
+  "steps": [
+    {
+      "id": "dimension-scope",
+      "name": "Dimension scope",
+      "description": "lock the competitor set + comparison dimensions",
+      "prompt": "1. Identify the subject (us/the product) and pick 3-6 named competitors that are genuinely comparable; justify each inclusion in one line. 2. Choose 5-8 comparison dimensions that decide the market (e.g. core features, pricing model, target segment, differentiator, weaknesses, traction signals) and define what each dimension means so cells are comparable. 3. Decide the matrix schema (rows = competitors incl. us, columns = dimensions) and where the final analysis lands. 4. Write an acceptance-criteria checklist (e.g. 'every competitor x every dimension cell is filled with evidence, not blank or guessed', 'our own offering is included as a row', 'each cell cites or names a source', 'the analysis ends with a recommendation and the top competitive gaps'). 5. Record decisions. Call write_task_note with the competitor set, dimensions, schema, and checklist; write the same to {{workPath}}/dimension-scope.md.\n\nThe deliverable `{{workPath}}/dimension-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": "{{workPath}}/dimension-scope.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/dimension-scope.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/dimension-scope.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "dimension-scope",
+        "maxAttempts": 3
+      },
+      "next": "research",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "artifact",
+        "additionalOutputMedia": [
+          "task-note"
+        ]
+      }
+    },
+    {
+      "id": "research",
+      "name": "Fill the matrix",
+      "description": "evidence-backed cell for every competitor x dimension",
+      "prompt": "1. For each competitor, research every locked dimension and fill the corresponding cell in {{workPath}}/matrix.md with a specific, evidence-backed value (a feature fact, a real price, a named segment) — never 'maybe' or a guess. 2. Cite or name the source for each cell (product page, pricing page, review, filing) in a [n] reference list. 3. Use consistent wording per dimension across competitors so cells are directly comparable. 4. Where a value is genuinely unknown, mark the cell UNKNOWN and say why, rather than fabricating. 5. Note any standout strength or weakness you observe per competitor for the writer to use. Good looks like a fully-populated, cited matrix. Do not write the narrative analysis yet.\n\nThe deliverable `{{workPath}}/matrix.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": "{{workPath}}/matrix.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/matrix.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/matrix.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "research",
+        "maxAttempts": 3
+      },
+      "next": "write",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "artifact"
+      }
+    },
+    {
+      "id": "write",
+      "name": "Write the analysis",
+      "description": "narrative analysis + recommendation from the matrix",
+      "prompt": "1. Write {{workPath}}/analysis.md: start with the rendered comparison matrix (a markdown table of competitors x dimensions), then a per-dimension or per-competitor narrative interpreting the matrix. 2. Explicitly call out where each competitor leads, where they are weak, and where the market has a gap. 3. End with a clear, defensible recommendation (where to compete, what to build/avoid) tied to specific cells in the matrix. 4. Keep every claim traceable to a matrix cell and its source; carry the [n] references through. 5. On a loop-back, fix only the named gaps. 6. Call write_task_note with the analysis path and which criteria now pass.\n\nThe deliverable `{{workPath}}/analysis.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}}/analysis.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/analysis.md",
+            "bytes": 1500,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/analysis.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "write",
+        "maxAttempts": 4
+      },
+      "next": "evaluate",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "artifact",
+        "additionalOutputMedia": [
+          "task-note"
+        ]
+      }
+    },
+    {
+      "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 {{workPath}}/analysis.md and check EACH criterion PASS/FAIL: (1) the matrix table covers every competitor x every dimension from the scope with no unexplained blanks; (2) our own offering is included; (3) cells are evidence-backed and consistently worded, with sources named (spot-check 3 against {{workPath}}/matrix.md); (4) the narrative actually interprets the matrix rather than restating it; (5) there is a clear recommendation tied to specific evidence; (6) competitive gaps are explicitly identified. A pretty write-up over a half-empty or guessed matrix FAILS.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/analysis.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: \"write\" })` 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}}/analysis.md",
+          "artifact": true
+        }
+      ],
+      "next": "write",
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "task-note"
+      }
+    },
+    {
+      "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,
+      "toolPolicy": {
+        "disallowBuiltinToolsets": [
+          "ai-apps",
+          "archives",
+          "artifacts",
+          "audio",
+          "browser-automation",
+          "code-execution",
+          "craftbooks",
+          "data-tables",
+          "entity-intel",
+          "git",
+          "image-intel",
+          "images",
+          "role-delegation",
+          "role-delegation-escalation",
+          "security-intel",
+          "team-management",
+          "videos",
+          "web",
+          "workspace-fs-write"
+        ],
+        "outputMedium": "task-note"
+      }
+    }
+  ],
+  "version": "1.0.3",
+  "releasedAt": "2026-09-01T20:00:00Z",
+  "minGezelVersion": "1.26233"
+}

package/data/craftbook-templates/co/competitive-analysis/versions/1.0.3/test.json

@@ -0,0 +1,119 @@
+{
+  "schemaVersion": 1,
+  "title": "Evidence-backed competitive matrix",
+  "objective": "Measure whether the competitive-analysis craftbook creates a complete cited matrix and a recommendation grounded in seeded competitor facts.",
+  "tags": [
+    "html-page"
+  ],
+  "prompt": "In the `Competitive Analysis Eval` project, read `source/competitor-briefs.md` and write `tasks/eval/analysis.md`: compare HarborCRM against AtlasOps, Deskline, and RelayFlow across target customer, pricing, automations, integrations, reporting, and weaknesses. Include a markdown comparison matrix, sources/citations from the seeded briefs, competitive gaps, and a recommendation for where HarborCRM should compete.",
+  "setup": {
+    "projectName": "Competitive Analysis Eval",
+    "about": "Self-contained eval project for competitive-analysis. All competitor facts are seeded in workspace/source/competitor-briefs.md.",
+    "missionObjectives": "Write a cited competitive analysis with a complete matrix, evidence-backed cells, competitive gaps, and a recommendation tied to the seeded facts.",
+    "files": [
+      {
+        "path": "source/competitor-briefs.md",
+        "content": "# Seeded competitor evidence\n\n[1] HarborCRM: Our product for regional wholesalers. Pricing is $149/user/month. Strengths: EDI order follow-up, inventory-aware reminders, native QuickBooks and NetSuite exports. Weaknesses: no built-in AI email summarization, reporting is CSV-first.\n\n[2] AtlasOps: Targets enterprise operations teams. Pricing starts at $40k/year platform fee plus seats. Strengths: workflow automations, approval routing, SSO, audit logs. Weaknesses: slow onboarding and limited wholesale-specific templates.\n\n[3] Deskline: Targets SMB support teams. Pricing is $79/user/month. Strengths: shared inbox, AI email summaries, helpdesk integrations, fast setup. Weaknesses: weak inventory context and no EDI order workflow.\n\n[4] RelayFlow: Targets logistics coordinators. Pricing is usage-based at $0.08/order event. Strengths: carrier tracking integrations, exception alerts, SLA dashboards. Weaknesses: not a CRM, limited account history, weak finance exports.\n"
+      }
+    ],
+    "worker": {
+      "name": "Iris",
+      "role": "Researcher"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "workspace/analysis.md contains a complete cited matrix, all four products, all dimensions, competitive gaps, and an evidence-tied recommendation.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/analysis.md",
+        "kind": "markdown-report",
+        "minBytes": 2400,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "HarborCRM|AtlasOps|Deskline|RelayFlow"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "target customer|pricing|automation|integration|reporting|weakness",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "\\|.*\\|.*\\|",
+            "flags": "m"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "\\[1\\]|\\[2\\]|\\[3\\]|\\[4\\]"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "competitive gap|gap|recommendation|compete",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "EDI|QuickBooks|NetSuite|AI email|carrier tracking",
+            "flags": "i"
+          },
+          {
+            "kind": "notContains",
+            "file": "tasks/eval/analysis.md",
+            "pattern": "TBD|unknown|not available",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/analysis.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "structure",
+        "description": "Layout and hierarchy fit the content; sections are scannable and coherent."
+      },
+      {
+        "name": "grounding",
+        "description": "Seeded source content is used faithfully — no invented facts or dropped requirements."
+      },
+      {
+        "name": "interactivity",
+        "description": "Controls and states the page claims to offer actually work."
+      },
+      {
+        "name": "polish",
+        "description": "Styling is intentional and responsive rather than unstyled or broken."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "evidence-backed cells",
+    "complete matrix",
+    "recommendation from facts"
+  ],
+  "extensions": {
+    "legacySimulators": [
+      {
+        "id": "seeded-competitor-corpus",
+        "kind": "data-source",
+        "status": "implemented",
+        "description": "Local evidence corpus replacing live web research for competitor facts."
+      }
+    ]
+  }
+}