@bendyline/gilde 0.1.38 → 0.1.39

package/data/craftbook-templates/so/social-thread/versions/1.0.2/craftbook.json

@@ -0,0 +1,119 @@
+{
+  "id": "social-thread",
+  "name": "Social Media Thread",
+  "description": "Write a multi-post social thread (X/Twitter, LinkedIn, Threads) that hooks, delivers, and converts: it first locks the single angle/big idea, the platform and its character limits, the target reader, and the CTA, then drafts the thread as numbered posts with a scroll-stopping hook, one idea per post, and a clear close. Covers Twitter thread, LinkedIn carousel copy, viral thread, social content, thought-leadership posts, and engagement hooks. Locking one angle and the hook before writing is the point: a small model that commits to a single big idea avoids the meandering thread that tries to say five things and stops no scroll.\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. Angle scope (planner) — lock the one big idea, platform limits, reader, and CTA → gated on artifact `{{workPath}}/angle-scope.md` (markdown-notes)\n2. Draft the thread (copywriter) — write the numbered thread to the locked angle and limits → gated on `social-thread.md` (markdown-doc)\n\nThe gates never advance with an unmet criterion, and loop back to the\nowning phase to fix named gaps.\n",
+  "entryStepId": "angle-scope",
+  "triggers": [
+    "write a twitter thread",
+    "social media thread",
+    "linkedin post",
+    "viral thread",
+    "write social content"
+  ],
+  "steps": [
+    {
+      "id": "angle-scope",
+      "name": "Angle scope",
+      "description": "lock the one big idea, platform limits, reader, and CTA",
+      "prompt": "1) State the ONE big idea/angle in a single sentence — the single takeaway the reader keeps. 2) Pick the platform and its constraints (e.g. X: <=280 chars/post; LinkedIn: <=3000 chars, line breaks matter). 3) Define the target reader and why this angle stops their scroll. 4) Decide the thread length (e.g. 6-10 posts) and the CTA for the final post (follow, link, reply prompt). 5) Sketch the post-by-post beat list: hook → 1 idea per post → close/CTA. Write an acceptance-criteria checklist (e.g. 'post 1 is a standalone scroll-stopping hook'; 'one idea per post'; 'every post within the char limit'; 'numbered and ordered'; 'final post has a clear CTA'; 'on a single consistent voice'). Use write_task_note for the angle + beats + checklist, and write the same to the produces path. No thread copy yet.\n\nThe deliverable `{{workPath}}/angle-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}}/angle-scope.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/angle-scope.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/angle-scope.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "angle-scope",
+        "maxAttempts": 3
+      },
+      "next": "draft"
+    },
+    {
+      "id": "draft",
+      "name": "Draft the thread",
+      "description": "write the numbered thread to the locked angle and limits",
+      "prompt": "Write the full thread as numbered posts. 1) Post 1 is the HOOK: a standalone, curiosity- or tension-driven opener that works even if no one reads the rest — no 'a thread:' throat-clearing as the only content. 2) Each subsequent post delivers exactly one idea from the beat list, building the argument. 3) Keep every post within the platform's character limit (count characters). 4) Use a punchy, consistent voice; short lines; concrete specifics over abstractions. 5) The final post delivers the CTA. Output markdown as a numbered list where each item is one post, with a character count noted in parentheses after each. Save to the produces path and note total post count.",
+      "suggestedRole": "copywriter",
+      "advanceWhen": {
+        "file": "social-thread.md",
+        "minBytes": 1,
+        "sniff": "nonempty"
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "social-thread.md",
+            "bytes": 500
+          }
+        ],
+        "scripts": [
+          {
+            "name": "checkContains",
+            "scope": "standard",
+            "inputs": {
+              "file": "social-thread.md",
+              "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+              "flags": "i"
+            }
+          }
+        ],
+        "onReject": "draft",
+        "maxAttempts": 4
+      },
+      "next": "evaluate"
+    },
+    {
+      "id": "evaluate",
+      "name": "Evaluate",
+      "description": "Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.",
+      "prompt": "Read social-thread.md and grade against the checklist. Check EACH and write PASS/FAIL: (a) post 1 is a standalone, scroll-stopping hook; (b) one idea per post; (c) every post is within the platform character limit (verify the counts); (d) posts are numbered and logically ordered; (e) the final post has a clear CTA; (f) a single consistent voice throughout. A thread that exceeds the char limit on any post or has a weak/missing hook FAILS — loop back to draft and name the posts to fix.\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: \"draft\" })` 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": "social-thread.md"
+        }
+      ],
+      "next": "draft"
+    },
+    {
+      "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
+    }
+  ],
+  "version": "1.0.2",
+  "releasedAt": "2026-08-21T00:00:00Z",
+  "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}}"
+      }
+    }
+  },
+  "minGezelVersion": "1.26233"
+}

package/data/craftbook-templates/so/social-thread/versions/1.0.2/test.json

@@ -0,0 +1,113 @@
+{
+  "schemaVersion": 1,
+  "title": "Social Media Thread smoke eval",
+  "objective": "Self-contained smoke eval for the Social Media Thread craftbook using the corpus generic harness.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "We need a social media thread 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": "Social Media Thread Eval",
+    "about": "Self-contained eval project for social-thread. Seeded inputs are under workspace/source or workspace/fixtures; final deliverable is workspace/report.md.",
+    "missionObjectives": "Use the Social Media Thread 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": "# Social Media Thread 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: social-thread - Social Media Thread.\n"
+      }
+    ],
+    "worker": {
+      "name": "Jules",
+      "role": "Writer"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "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"
+  ]
+}

package/data/craftbook-templates/so/source-quality-audit/versions/1.0.2/craftbook.json

@@ -0,0 +1,152 @@
+{
+  "id": "source-quality-audit",
+  "name": "Source Quality Audit",
+  "description": "Once a month, make the reading list earn its place: score each source on what it actually delivered — accuracy, originality, noise — and keep, demote, or drop with evidence, so the brief stays sharp as the feeds drift. Locks the audit criteria FIRST, then builds the per-source scorecard from the month's briefs, then writes the verdict table with the changes to make. Use for source audits, pruning the feed list, 'the brief feels noisy', monthly reading-list reviews, and deciding which feeds to drop.\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. Scope the audit (reviewer) — roster read, evidence base gathered, verdict rules locked → gated on artifact `{{workPath}}/scope.md` (markdown-notes)\n2. Build the scorecard (reviewer) — per-source evidence: delivered, corrected, or silent → gated on artifact `{{workPath}}/audit.md` (markdown-notes)\n3. Write the audit report (copywriter) — verdict table plus the changes to make → gated on artifact `{{workPath}}/report.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": "scope",
+  "triggers": [
+    "audit the sources",
+    "the brief feels noisy",
+    "review the reading list",
+    "which feeds should we drop",
+    "source quality audit"
+  ],
+  "steps": [
+    {
+      "id": "scope",
+      "name": "Scope the audit",
+      "description": "roster read, evidence base gathered, verdict rules locked",
+      "prompt": "Define the audit before scoring anything. Step 1: read sources.json from the workspace — every source on the roster — and gather the month's briefs from briefs/ as the evidence base. Step 2: lock the audit criteria — every source on the roster gets a verdict; verdicts come from a fixed set (keep, demote, drop); and every verdict must cite concrete evidence from the briefs: items it delivered, corrections it caused, or the silence of never being used. Step 3: agree what each verdict means in practice — keep means earning its place, demote means probation with a named reason, drop means off the list now. Step 4: write an acceptance-criteria checklist ('every source in sources.json appears in the verdict table', 'every verdict is keep, demote, or drop', 'every verdict cites at least one concrete example or explicitly notes the source went unused', 'the report ends with the changes to make'). Call write_task_note with the source count + criteria + checklist and write {{workPath}}/scope.md.\n\nThe deliverable `{{workPath}}/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": "reviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/scope.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/scope.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/scope.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "scope",
+        "maxAttempts": 3
+      },
+      "next": "audit"
+    },
+    {
+      "id": "audit",
+      "name": "Build the scorecard",
+      "description": "per-source evidence: delivered, corrected, or silent",
+      "prompt": "Build the per-source scorecard from the month's evidence. Step 1: for each source in the roster, sweep the month's briefs and collect what it actually delivered — items attributed to it, stories it broke versus merely echoed, corrections or walk-backs it caused, and whether it appeared at all. Step 2: score each source in plain terms — accuracy (was it right), originality (was it first or just loud), and noise (how much of it was worth the owner's morning). Step 3: draft a verdict per source from the fixed set with the evidence beside it. Step 4: write the per-source scorecard with cited examples to {{workPath}}/audit.md and call write_task_note with the draft verdict counts.\n\nThe deliverable `{{workPath}}/audit.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": "reviewer",
+      "advanceWhen": {
+        "file": "{{workPath}}/audit.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/audit.md",
+            "bytes": 120,
+            "artifact": true
+          },
+          {
+            "kind": "sniff",
+            "file": "{{workPath}}/audit.md",
+            "sniff": "nonempty",
+            "artifact": true
+          }
+        ],
+        "onReject": "audit",
+        "maxAttempts": 3
+      },
+      "next": "report"
+    },
+    {
+      "id": "report",
+      "name": "Write the audit report",
+      "description": "verdict table plus the changes to make",
+      "prompt": "Write the audit report the owner acts on. Step 1: open with the month in one honest paragraph — what the reading list did well and where the noise came from. Step 2: the verdict table — a Markdown table with columns Source, Verdict, and Why, one row per source, every verdict backed by its best evidence in a phrase. Step 3: the changes to make — which sources to drop or demote now, and what that leaves covered or exposed. Step 4: note anything worth adding to the roster to fill a gap the audit exposed. Keep it fair and specific — sources earn their place with evidence, not vibes. Write {{workPath}}/report.md and call write_task_note with the keep/demote/drop counts.\n\nThe deliverable `{{workPath}}/report.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}}/report.md",
+        "minBytes": 1,
+        "sniff": "nonempty",
+        "artifact": true
+      },
+      "gate": {
+        "at": "completion",
+        "checks": [
+          {
+            "kind": "minBytes",
+            "file": "{{workPath}}/report.md",
+            "bytes": 1200,
+            "artifact": true
+          },
+          {
+            "kind": "contains",
+            "file": "{{workPath}}/report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i",
+            "label": "at least one markdown heading",
+            "artifact": true
+          }
+        ],
+        "onReject": "report",
+        "maxAttempts": 4
+      },
+      "next": "evaluate"
+    },
+    {
+      "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}}/report.md, {{workPath}}/audit.md, and the month's briefs and verify every criterion from {{workPath}}/scope.md: every source in sources.json has a row in the verdict table, every verdict is keep, demote, or drop, every verdict cites at least one concrete example from the briefs (spot-check two against the brief files) or explicitly notes the source went unused, and the closing changes list matches the verdicts. Write PASS/FAIL per criterion; on any failure, name the gap and loop back to report.\n\nThe deliverable lives in the project's artifacts drawer — open `{{workPath}}/report.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: \"report\" })` 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}}/report.md",
+          "artifact": true
+        }
+      ],
+      "next": "report"
+    },
+    {
+      "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
+    }
+  ],
+  "version": "1.0.2",
+  "releasedAt": "2026-08-21T00:00:00Z",
+  "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}}"
+      }
+    }
+  },
+  "minGezelVersion": "1.26233"
+}

package/data/craftbook-templates/so/source-quality-audit/versions/1.0.2/test.json

@@ -0,0 +1,161 @@
+{
+  "schemaVersion": 1,
+  "title": "Source Quality Audit smoke eval",
+  "objective": "Task-specific eval for the Source Quality Audit craftbook: five roster sources against a month of briefs in which The Daily Splash caused two corrections and Materials Weekly never appeared — the verdict table covers all five with evidence.",
+  "tags": [
+    "corpus"
+  ],
+  "prompt": "the brief has felt noisy latly — can you audit the sources this month and tell me what to drop?",
+  "setup": {
+    "projectName": "Topic Watch — Source Audit Eval",
+    "about": "Self-contained eval project for source-quality-audit. The source roster is seeded at sources.json and the month's briefs sit under briefs/ as the evidence base; the deliverable is tasks/eval/report.md with the verdict table.",
+    "missionObjectives": "Use the Source Quality Audit craftbook/template, score every roster source against the month's briefs only, and write tasks/eval/report.md. No live feeds, services, or outside data — the seeded briefs are the whole record.",
+    "files": [
+      {
+        "path": "sources.json",
+        "content": "[\n  {\n    \"id\": \"newsline\",\n    \"name\": \"Newsline\",\n    \"kind\": \"regional news\",\n    \"added\": \"2026-01-12\"\n  },\n  {\n    \"id\": \"capitol-wire\",\n    \"name\": \"Capitol Wire\",\n    \"kind\": \"state policy wire\",\n    \"added\": \"2026-02-03\"\n  },\n  {\n    \"id\": \"harbor-bulletin\",\n    \"name\": \"Harbor Bulletin\",\n    \"kind\": \"local bulletin\",\n    \"added\": \"2026-03-20\"\n  },\n  {\n    \"id\": \"daily-splash\",\n    \"name\": \"The Daily Splash\",\n    \"kind\": \"aggregator\",\n    \"added\": \"2026-05-08\"\n  },\n  {\n    \"id\": \"materials-weekly\",\n    \"name\": \"Materials Weekly\",\n    \"kind\": \"trade journal\",\n    \"added\": \"2026-04-14\"\n  }\n]\n"
+      },
+      {
+        "path": "briefs/2026-06-05.md",
+        "content": "# Morning brief - 2026-06-05\n\n## Municipal Broadband\n- Council schedules the broadband vote for July. (Newsline)\n- The Daily Splash claims the depot fiber build cut program costs 40%. (The Daily Splash)\n\n## Battery Recycling\n- Spring collection pilot wraps with volumes above plan. (Harbor Bulletin)\n"
+      },
+      {
+        "path": "briefs/2026-06-12.md",
+        "content": "# Morning brief - 2026-06-12\n\n## Municipal Broadband\n- Correction: the 40% cost figure The Daily Splash ran last week was wrong - the county puts the saving at 14%. (Harbor Bulletin)\n- State fiber office publishes the pilot application checklist. (Capitol Wire)\n\n## Battery Recycling\n- Port recycler files for the second-life pack permit. (Harbor Bulletin)\n"
+      },
+      {
+        "path": "briefs/2026-06-19.md",
+        "content": "# Morning brief - 2026-06-19\n\n## Municipal Broadband\n- Newsline breaks the story: the cost study's consultant missed two substations, delaying the vote. (Newsline)\n- The Daily Splash reported the Westport plant had closed; the plant says otherwise - treating as retracted. (The Daily Splash)\n\n## Battery Recycling\n- Fire marshal opens comment period on depot storage rules. (Capitol Wire)\n"
+      },
+      {
+        "path": "briefs/2026-06-26.md",
+        "content": "# Morning brief - 2026-06-26\n\n## Municipal Broadband\n- Council staff confirm the revised cost study lands mid-July. (Newsline)\n\n## Battery Recycling\n- Weekend depots extend hours for the summer. (Harbor Bulletin)\n- The Daily Splash runs six items on a viral battery-fire video; none add facts beyond the original report. (The Daily Splash)\n"
+      }
+    ],
+    "worker": {
+      "name": "Isa",
+      "role": "Editor"
+    },
+    "craftbookParams": {
+      "workPath": "tasks/eval"
+    }
+  },
+  "mocks": [],
+  "success": {
+    "summary": "tasks/eval/report.md carries a verdict table covering all five roster sources, verdicts from the fixed set with cited evidence: The Daily Splash's corrections and Materials Weekly's silence both show up in their rows.",
+    "deliverables": [
+      {
+        "path": "tasks/eval/report.md",
+        "kind": "markdown-report",
+        "minBytes": 900,
+        "checks": [
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "(?:^|\\n)#{1,3}\\s+\\S",
+            "flags": "i"
+          },
+          {
+            "kind": "tableShape",
+            "file": "tasks/eval/report.md",
+            "requiredColumns": [
+              "Source",
+              "Verdict"
+            ],
+            "minRows": 5
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "keep",
+            "flags": "i",
+            "label": "verdicts use the fixed set — keep"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "drop",
+            "flags": "i",
+            "label": "verdicts use the fixed set — drop"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "Newsline",
+            "flags": "i",
+            "label": "every roster source covered — Newsline"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "Capitol Wire",
+            "flags": "i",
+            "label": "every roster source covered — Capitol Wire"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "Harbor Bulletin",
+            "flags": "i",
+            "label": "every roster source covered — Harbor Bulletin"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "Daily Splash",
+            "flags": "i",
+            "label": "every roster source covered — The Daily Splash"
+          },
+          {
+            "kind": "contains",
+            "file": "tasks/eval/report.md",
+            "pattern": "Materials Weekly",
+            "flags": "i",
+            "label": "every roster source covered — Materials Weekly"
+          },
+          {
+            "kind": "judge",
+            "file": "tasks/eval/report.md",
+            "rubric": "Every verdict cites at least one concrete example from the month's briefs or explicitly notes the source went unused. The Daily Splash's repeated corrections and Materials Weekly's total silence are both reflected in their verdicts.",
+            "sourceFiles": [
+              "briefs/2026-06-05.md",
+              "briefs/2026-06-12.md",
+              "briefs/2026-06-19.md",
+              "briefs/2026-06-26.md"
+            ],
+            "label": "verdicts earn their evidence"
+          }
+        ]
+      }
+    ]
+  },
+  "rubric": {
+    "artifact": {
+      "path": "tasks/eval/report.md",
+      "kind": "markdown"
+    },
+    "axes": [
+      {
+        "name": "evidence",
+        "description": "Every verdict is backed by concrete examples from the briefs — items delivered, corrections caused, or documented silence."
+      },
+      {
+        "name": "fairness",
+        "description": "Sources are judged on the record, not vibes; a noisy source's genuine contributions still count."
+      },
+      {
+        "name": "decisiveness",
+        "description": "Verdicts come from the fixed set (keep, demote, drop) and the closing changes list matches them."
+      },
+      {
+        "name": "hygiene",
+        "description": "The report notes what dropping a source leaves uncovered and what might fill the gap."
+      }
+    ]
+  },
+  "qualityFocus": [
+    "evidence-backed verdicts",
+    "full roster coverage",
+    "actionable changes"
+  ]
+}