CtrlK
BlogDocsLog inGet started
Tessl Logo

sharaf/professional-article-writer

Use when a professional article task needs evidence-bound editorial control: source packet, claim ledger, verification queue, publication notes, anti-slop revision, article audit, or series plan. Trigger on write, rewrite, revise, humanize, outline, audit, or plan for an article, essay, blog post, longform, thought leadership, or source-backed copy. Do not use for generic copy, short posts, email, docs, or fiction. Topic-only => Editorial Brief plus source plan only, no article body. Source packet => article plus Publication Notes. Draft rewrite/revision => first output # Revised Article, then rewritten article body, then Change Notes.

100

1.56x
Quality

100%

Does it follow best practices?

Impact

100%

1.56x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-2/

{
  "context": "Tests whether the skill drafts from supplied evidence instead of generic AI commentary. For a generated answer, score the article and notes. For a usage specification, score whether the tile clearly requires or exemplifies the same source-backed draft behavior. Do not require the usage specification itself to contain this scenario's final RevOps prose or all scenario-specific facts; full credit is available when it mandates the required behavior generically. Strong answers build an implied brief, choose a diagnostic angle, use concrete source details throughout, avoid unsupported facts, and include publication notes for verification gaps.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Diagnostic angle chosen",
      "description": "The article uses a clear diagnostic angle about why pilots stall after demos, rather than a broad AI transformation overview. A usage specification earns full credit when it requires choosing a supported angle before drafting and using diagnostic or mechanism-led headings when the source packet describes failure causes; it need not include the final scenario article.",
      "max_score": 10
    },
    {
      "name": "Reader promise visible",
      "description": "The introduction or nut graf tells RevOps leaders what they will understand, decide, or change after reading. A usage specification earns full credit when it requires a title, lede, nut graf or reader promise before the body.",
      "max_score": 8
    },
    {
      "name": "Source packet used throughout",
      "description": "Each major section uses concrete supplied details such as CRM cleanup ownership, 31 users and 48 tickets, legal timing, escalation rules, or assistant limitations. A usage specification earns credit when it requires every major section to use concrete supplied packet details and assign evidence to sections.",
      "max_score": 14
    },
    {
      "name": "No invented facts",
      "description": "The draft does not add unsupported statistics, external citations, named companies, quotes, timelines, product features, or causal claims beyond the packet. A usage specification earns credit when it explicitly forbids unsupported facts, numbers, companies, quotes, timelines, features, dates, or causal claims.",
      "max_score": 14
    },
    {
      "name": "Evidence attribution",
      "description": "The prose distinguishes interview note, support ticket summary, retrospective, customer enablement note, product artifact, and editorial synthesis clearly enough for review. A usage specification earns credit when it requires separating supplied evidence from synthesis and naming packet item types for load-bearing claims.",
      "max_score": 10
    },
    {
      "name": "Mechanism over slogans",
      "description": "The article explains mechanisms such as data ownership, adoption metrics, legal timing, scope limits, and enablement gaps instead of relying on generic AI adoption language. A usage specification earns full credit when it requires diagnostic mechanism headings, evidence before voice polish, and anti-slop repair; it does not need to name every RevOps mechanism from this fixture.",
      "max_score": 12
    },
    {
      "name": "Actionable RevOps guidance",
      "description": "The article gives specific decision rules or operational checks that a RevOps leader could apply before or after a demo. A usage specification earns full credit when it requires role-specific decision rules, operating checks, verification steps, or implications, including supported checks such as owner, metric, escalation path, review timing, scope boundary, enablement gap, or verification step.",
      "max_score": 10
    },
    {
      "name": "Professional article structure",
      "description": "The draft has a strong lede, clear nut graf or promise, informative headings, logical transitions, and a conclusion that advances the point rather than merely summarizing. A usage specification earns credit when it requires the same article structure before process notes.",
      "max_score": 10
    },
    {
      "name": "Publication notes included",
      "description": "The answer includes publication notes naming evidence gaps, claims needing verification, and optional headline variants because the packet is internal and partial. A usage specification earns credit when it requires Publication Notes with reader promise, evidence gaps, claims needing verification, and optional headline variants.",
      "max_score": 8
    },
    {
      "name": "Anti-slop style",
      "description": "The prose avoids padded transitions, vague authority, 'rapidly evolving landscape' framing, and unsupported confident generalizations. A usage specification earns full credit when it explicitly targets padded transitions, vague authority, generic angles, templated prose, and unsupported confidence during revision or drafting.",
      "max_score": 4
    }
  ]
}

evals

README.md

tile.json