Long-Form Article Writer with Citations
The gold standard. Turn a research brief into a 2,000-4,000 word article with inline citations. Refuses to invent sources. Brand voice configurable. Best-in-class citation discipline.
Every AI Skill in our directory that produces written content for the outside world — blog posts, newsletters, marketing copy, SEO meta tags, social posts, thought-leadership pieces. Each Skill has been used against real briefs by our editorial team and documented for Claude, ChatGPT, Cursor, and Cline. Every Skill enforces citation discipline where relevant — nothing invented, everything traceable.
Content Creation Skills accept a brief (topic, audience, tone, length, source material) and produce written content — an article, a newsletter section, a landing-page block, a social post, a meta tag. The best of these Skills preserve voice (yours or your brand's), respect citations when working from source material, and enforce discipline around length limits (SEO title/description char counts, tweet limits, newsletter word budgets).
The value the Skill layer adds is consistency. Plain LLM output for content work drifts from your voice, forgets citation discipline mid-draft, and quietly exceeds length limits. Skills lock in the constraints — voice samples for tone, source-only citation rules, character-count validation before returning. For content teams, this is the difference between AI drafts you edit lightly and AI drafts you rewrite from scratch.
What lives here vs. elsewhere: if the output is written content for external distribution, it belongs here. If it's an internal document with formatting (a Word doc, a formal report), see Document Creation. If it's the research work that precedes the writing, see Research & Analysis. If it's design assets that accompany content (banners, illustrations), see Design & Visual.
Our editorial picks — heaviest tested, highest reader ratings.
The gold standard. Turn a research brief into a 2,000-4,000 word article with inline citations. Refuses to invent sources. Brand voice configurable. Best-in-class citation discipline.
Character-validated meta tags. Titles under 60 chars, descriptions under 155. Keyword-aware and click-worthy. Bulk mode processes CMS exports in one pass.
Draft your weekly newsletter in your voice from a set of source items. Learns from your previous issues. Enforces your word budget. Includes a subject-line generator.
Filter by sub-type, search by name, sort by recency or alphabetical.
Turn a research brief into a 2,000-4,000 word article with inline citations. Refuses to invent sources. Brand voice configurable.
Given an interview transcript, produce a structured customer case study (situation, approach, results). Preserves quotes verbatim.
Draft 5,000-8,000 word whitepapers from a research brief. Formal register, executive summary, methodology section.
Plan a multi-chapter eBook from a topic + audience + goal. Drafts each chapter iteratively with revision passes.
Generate a well-structured blog post outline from a target keyword and rough angle. Sub-heads, key points per section, suggested word count.
Generate 5-10 title variants for A/B testing. Different angles: benefit, curiosity, listicle, contrarian, question.
Edit a draft against a house style guide. Suggests changes with tracked-changes-style annotations for review.
Given a pillar topic, plan a topic cluster of 8-15 related posts with internal linking structure.
Draft weekly newsletter in your voice from source items. Learns from previous issues. Enforces word budget.
Generate open-rate-optimized subject lines. Multiple variants with different psychological hooks. Character-count validated.
Summarize newsletter analytics (opens, clicks, unsubs) into a plain-English brief highlighting what to change.
Character-validated meta tags. Titles under 60 chars, descriptions under 155. Keyword-aware. Bulk mode for CMS exports.
Generate JSON-LD schema markup (Article, HowTo, FAQPage, Product) from page content. Google-Rich-Results-Test valid.
Generate a brief for a freelancer or in-house writer from a target keyword. Includes competitor analysis and outline.
Given a new post, scan existing posts for good internal link candidates. Suggests anchor text and target URLs.
Generate X/Twitter posts and threads. Character-aware, hook-first, thread-structured. Voice-matched to your handle.
LinkedIn-native posts with the platform's specific tone. Longer than Twitter, story-driven, first-person.
Given a blog post, generate 5-10 social posts (Twitter, LinkedIn, Threads) that promote it without duplicating.
Generate landing page copy blocks (headline, sub, features, CTA) from a product spec. Conversion-focused patterns.
Draft a 5-7 email nurture sequence for a specific offer. Each email has a job in the sequence (welcome, education, offer).
Generate product descriptions from specs. Multiple length variants (short, medium, long). SEO-optimized.
MCP servers our team pairs with Content Creation Skills for full workflows.
Publish drafts to WordPress. Pairs with every long-form and blog Skill for full drafting-to-publishing workflows.
Save drafts to a Notion editorial workspace for team review before publishing.
Draft in Google Docs for shared editing. Read source materials from Drive.
Research backend for citation-heavy long-form work. Pairs with the Long-Form Article Writer Skill.
Send newsletter drafts to yourself for phone-based review before final send.
Announce new content to team channels or seek in-thread editorial review.
Longer-form tutorials for the concepts behind these Skills.
How to build a voice-guide file that gets AI to sound consistently on-brand.
The specific techniques that keep AI-drafted articles source-accurate rather than plausibly-worded fiction.
How search is changing and how content strategy is adapting.
Head-to-head reviews that touch on Content Creation workflows.
Head-to-head on 20 real content briefs. Verdict: Claude ahead on voice consistency, GPT-5 ahead on tight structure.
Which search backend produces the most usable source material for long-form work?
In-product AI vs Skills-based drafting: when each wins.
voice-guide.md file you provide once. The guide specifies tone, hedging language preferences, banned phrases, preferred patterns, plus 2-3 sample paragraphs of on-brand writing. Skills use those samples as few-shot examples during generation. This is much more effective than describing your voice in prose — the model reads the samples and matches them. Voice adherence is best on Claude Opus 4.7 and GPT-5, both of which handle few-shot style transfer reliably.[NEEDS SOURCE — editor review] flag inline, or (c) the claim dropped. Fabrication rate in our test corpus is 0.1% for the Long-Form Article Writer Skill. See its dedicated page for the full test methodology.Our sister sites cover errors and code recipes across the AI ecosystem. Bookmark them for when a Skill hits an edge case.
Weekly roundup of new Skills we've tested, install notes, and workflow patterns. Join 2,400+ readers of "The Skill Report" every Tuesday.