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
100%
Does it follow best practices?
Impact
100%
1.56xAverage score across 3 eval scenarios
Passed
No known issues
task.mdevals/scenario-2/
Source-Backed Draft Request
Write a professional article for RevOps leaders about why internal AI pilots stall after the demo stage. Use the supplied source packet. The article should be concrete, useful, and publication-ready, but include publication notes if anything still needs verification.
Source packet:
- Interview note, VP Revenue Operations, May 2026: "The prototype answered questions well in a sandbox, but nobody owned the CRM field cleanup, so the answers looked impressive until reps tried real accounts."
- Support ticket summary, April 2026: 31 pilot users opened 48 tickets in two weeks. The three most common issues were missing account context, unclear escalation rules, and managers asking for exports that were not in scope.
- Internal pilot retrospective: The team measured demo completion but did not define a production adoption metric. Legal review started after the demo, not before the pilot.
- Customer enablement note: Frontline reps wanted examples of when not to use the assistant, not only a launch announcement.
- Product artifact: The assistant can summarize account notes, draft follow-up emails, and flag missing fields. It cannot update CRM records or approve discount exceptions.
Draft about 900 words. Avoid generic AI-transformation framing.