@bendyline/gilde 0.1.51 → 0.1.53

package/data/craftbook-templates/mi/micro-game-jam/versions/1.0.3/test.json

@@ -0,0 +1,99 @@
+{
+  "schemaVersion": 1,
+  "title": "Micro Game Jam smoke eval",
+  "objective": "Self-contained smoke eval for the Micro Game Jam craftbook using the html-game generic harness.",
+  "tags": [
+    "html-game"
+  ],
+  "prompt": "Make us a fun litle browser game around the returns-desk theme — just one file, index.html, that we can open and play.",
+  "setup": {
+    "projectName": "Micro Game Jam Eval",
+    "about": "Self-contained eval project for micro-game-jam. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/index.html.",
+    "missionObjectives": "Use the Micro Game Jam craftbook/template, read the seeded local fixtures, and write index.html without network calls, real credentials, or live services.",
+    "files": [
+      {
+        "path": "source/brief.md",
+        "content": "# Micro Game Jam 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: micro-game-jam - Micro Game Jam.\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Developer"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "index.html is a complete self-contained browser game smoke artifact.",
+    "deliverables": [
+      {
+        "path": "index.html",
+        "kind": "html-game",
+        "minBytes": 3200,
+        "checks": [
+          {
+            "kind": "cssMinBytes",
+            "bytes": 600,
+            "file": "index.html"
+          },
+          {
+            "kind": "jsParses",
+            "file": "index.html"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "<canvas|<svg|requestAnimationFrame",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "score|state|level|turn",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "restart|play again|new game|game over|victory|you win|you lose",
+            "flags": "i"
+          },
+          {
+            "kind": "contains",
+            "file": "index.html",
+            "pattern": "Boreal|returns|desk",
+            "flags": "i"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "index.html",
+      "kind": "html"
+    },
+    "axes": [
+      {
+        "name": "playability",
+        "description": "The game is actually playable: input works, state advances, and a session can be completed."
+      },
+      {
+        "name": "completeness",
+        "description": "Core loop, scoring/win-lose, and restart are all present rather than stubbed."
+      },
+      {
+        "name": "polish",
+        "description": "Visual design and feedback (motion, states, messages) feel deliberate, not default."
+      },
+      {
+        "name": "robustness",
+        "description": "No console errors, dead controls, or states the player can get stuck in."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "runtime surface",
+    "interaction loop",
+    "self-contained game"
+  ]
+}

package/data/craftbook-templates/mo/mock-interview/versions/1.0.3/craftbook.json

@@ -0,0 +1,270 @@
+{
+  "id": "mock-interview",
+  "name": "Mock Interview",
+  "description": "A realistic practice interview run live with the candidate — planned questions calibrated to the target role and round, a real-time interview with pressure follow-ups, and a kind, specific debrief: strengths in the candidate's own words, fixes with sketched stronger answers, one drill, and an honest readiness read. The debrief is the deliverable. Use before any real interview round.\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. Plan the interview (Oefen-interviewer) — six to ten calibrated questions with planned follow-up chains → gated on artifact `{{workPath}}/interview-plan.md` (markdown-notes)\n2. Conduct the interview (Oefen-interviewer) — the live practice round, transcribed as it runs → gated on artifact `{{workPath}}/interview-transcript.md` (markdown-notes)\n3. Write the debrief (Oefen-interviewer) — kind, specific, grounded in the transcript → gated on artifact `{{workPath}}/reports/interview-debrief.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": "plan",
+  "triggers": [
+    "mock interview",
+    "practice interview",
+    "interview prep session",
+    "run me through a practice round"
+  ],
+  "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": "plan",
+      "name": "Plan the interview",
+      "description": "six to ten calibrated questions with planned follow-up chains",
+      "prompt": "Build the interview plan. Step 1: establish the target role, company, and round — read the project's pipeline record and company brief when present; otherwise ask, briefly. Step 2: draft six to ten questions mixing behavioral and role-specific, calibrated to the round (a recruiter screen is not a final loop). Step 3: for each major expected claim, plan a follow-up chain — the 'what did YOU do?' probe. At least one chain must apply real pressure. Step 4: define the debrief contract up front: two or three strengths quoting the candidate, two or three fixes each with a sketched stronger answer, exactly one drill, and a readiness read. Step 5: write an acceptance-criteria checklist ('questions specific to the target role', 'at least one pressure follow-up chain', 'debrief contract stated'). Write the plan + checklist via write_task_note AND to {{workPath}}/interview-plan.md.\n\nThe deliverable `{{workPath}}/interview-plan.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": "Oefen-interviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/interview-plan.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/interview-plan.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/interview-plan.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "plan",
+        "maxAttempts": 3
+      },
+      "next": "conduct",
+      "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": "conduct",
+      "name": "Conduct the interview",
+      "description": "the live practice round, transcribed as it runs",
+      "prompt": "Run the interview live with the user. Step 1: set the frame in one line ('I'm your interviewer for a <round> for <role> — ready?') then go in character. Step 2: one question per turn, through the user-interaction capabilities on your current roster; wait for the answer before the next. Step 3: use the planned follow-up chains on vague or team-credited answers; stay realistic, never cruel — if the candidate is rattled, step out of character, steady them, resume when ready. Step 4: after each exchange, append a condensed record to {{workPath}}/interview-transcript.md: the question, the gist of the answer, your one-line read. Step 5: end the roleplay explicitly ('Stepping out of character now') once the planned questions or the candidate's energy are spent. The transcript file is the phase's deliverable — keep it current as you go, not reconstructed after.\n\nThe deliverable `{{workPath}}/interview-transcript.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": "Oefen-interviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/interview-transcript.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/interview-transcript.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/interview-transcript.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "conduct",
+        "maxAttempts": 3
+      },
+      "next": "debrief",
+      "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"
+      }
+    },
+    {
+      "id": "debrief",
+      "name": "Write the debrief",
+      "description": "kind, specific, grounded in the transcript",
+      "prompt": "Write {{workPath}}/reports/interview-debrief.md to the contract from the plan. Step 1: '## Strengths' — two or three, each quoting the candidate's own words from the transcript so they can hear what worked. Step 2: '## Fixes' — two or three, each naming the moment it surfaced, why it lands badly in a real round, and a sketched stronger answer they could actually give. Step 3: '## Drill' — exactly one exercise to run before the real interview, matched to the biggest fix. Step 4: '## Readiness' — an honest read: ready, or one more round (and on what). Step 5: every point must trace to the transcript; no generic interview advice. On a loop-back, fix ONLY the named issues.\n\nThe deliverable `{{workPath}}/reports/interview-debrief.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": "Oefen-interviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/reports/interview-debrief.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/reports/interview-debrief.md",
+            "bytes": 1500,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/reports/interview-debrief.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "debrief",
+        "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"
+      }
+    },
+    {
+      "id": "evaluate",
+      "name": "Evaluate",
+      "description": "Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.",
+      "prompt": "Review the debrief as the candidate's coach. Step 1: cross-check every fix and strength against {{workPath}}/interview-transcript.md — FAIL any point that does not trace to an actual exchange. Step 2: check the tone — kind and concrete; FAIL generic advice ('be more confident') that names no moment and sketches no stronger answer. Step 3: confirm exactly one drill, actionable tonight. Step 4: confirm the readiness read is honest about whether another round is needed. Write PASS/FAIL per criterion.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/reports/interview-debrief.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: \"debrief\" })` 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}}/reports/interview-debrief.md",
+          "artifact": true
+        }
+      ],
+      "next": "debrief",
+      "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/mo/mock-interview/versions/1.0.3/test.json

@@ -0,0 +1,110 @@
+{
+  "schemaVersion": 1,
+  "title": "Mock Interview smoke eval",
+  "objective": "Self-contained smoke eval for the Mock Interview craftbook using the corpus generic harness.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "We need a mock interview 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": "Mock Interview Eval",
+    "about": "Self-contained eval project for mock-interview. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/report.md.",
+    "missionObjectives": "Use the Mock Interview 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": "# Mock Interview 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: mock-interview - Mock Interview.\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"
+  ]
+}