agentleFS
timscheueraitimscheuerai/content-vault/skills/long-form/SKILL.md

long-form

Skill38 starsChanged 6 months agoMITView on GitHub

Draft written long-form content (newsletter issue, blog post, Substack issue) in your voice. 800-2000 words, structured intro + body + CTA, paragraph-by-paragraph not list-by-list. Pulls voice from /linkedin-copywriter's corpus (same author, same register · longer arc). Runs the brand-doc skill checklist before output. Logs the row in the Notion Content DB and (for newsletters) drafts a separate punchy subject line. Use when asked to "write a newsletter", "draft a blog post", "long-form X", "Substack issue about Y", "weekly newsletter", or "/long-form". For spoken video scripts (YouTube longs / shorts) use /youtube-script instead · the cadence is fundamentally different.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 8 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Target length: default 1200-1500 words (newsletter), 800-1500 (blog)

    line 38
  2. Distinctive:

    If anything is missing, ask in one batch, then draft.

    line 46
  3. Distinctive:

    before drafting: if you've been off-task. Top 5 posts are the strongest voice anchors.

    line 53
  4. Distinctive:

    If you find yourself padding to hit length, the topic isn't deep enough.

    line 64
  5. Distinctive:

    All caps · spam folder

    line 146
  6. Distinctive:

    Run the brand-doc skill checklist on the final draft.

    line 152
  7. Distinctive:

    If any check fails, fix and re-run.

    line 162
  8. Very specific:

    Create a row via mcp__claude_ai_Notion__notion-create-pages with data source <YOUR_NOTION_CONTENT_DB_ID>:

    line 170
  9. Very specific:

    If invoked from /repurpose, the orchestrator may handle row creation · check the source context to avoid duplication.

    line 200
  10. Distinctive:

    Don't pad to hit length: If 1500 words feels thin, propose a shorter shape (LinkedIn post or thread) and stop.

    line 216
  11. Very specific:

    Don't fabricate quotes / metrics / customer names: If you're drawing from /repurpose source or a customer-interview transcript, use exact …

    line 233

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

long-form

Written long-form content. One author voice (yours), three default shapes (newsletter / blog / Substack issue). Same skill, different register and length per shape.

This is written long-form. For spoken long-form (YouTube longs, podcast monologues, talking-head video scripts), use /youtube-script · spoken cadence is structurally different and trying to merge them produces worse output for both.

When to use

Trigger on:

  • "Write a newsletter issue"
  • "Draft a blog post about X"
  • "Long-form / Substack issue on Y"
  • "Weekly newsletter"
  • "Article for the website"
  • User explicitly types /long-form

Skip for:

  • LinkedIn / X posts → /linkedin-copywriter · /x-copywriter
  • YouTube / podcast / talking-head scripts → /youtube-script
  • Lead magnet PDFs → /lead-magnet-creator
  • Sales proposals → /proposal-creator

Inputs

The brief should include:

  • Type: newsletter (default) · blog · substack-issue
  • Topic + angle (one sentence)
  • Target length: default 1200-1500 words (newsletter), 800-1500 (blog)
  • Audience: who's reading? (newsletter list, cold blog visitor, both)
  • Optional source: Notion row URL if invoked from /repurpose · this anchors the piece to a specific master and informs the body
  • Optional CTA: what do you want the reader to do at the end? · default: "reply to this email" (newsletter), "subscribe to the newsletter" (blog), "tweet me" (substack)

If anything is missing, ask in one batch, then draft.

Voice anchor

The voice is yours, the same as /linkedin-copywriter. Read the linkedin-copywriter corpus you've populated (../linkedin-copywriter/corpus.md) before drafting if you've been off-task. Top 5 posts are the strongest voice anchors.

Long-form differs from short-form in arc and density, not register:

  • Arc: short-form is one beat (hook + payoff). Long-form is 3-5 beats. You can develop a thought.
  • Density: short-form is compressed (every line earns its place). Long-form has breathing room (paragraphs, not bullet-stacks).
  • Voice: identical. Same opinions, same word choices, same no-em-dash / no-slop discipline.

If you find yourself padding to hit length, the topic isn't deep enough. Tell the user, propose a shorter shape (LinkedIn post or thread), don't force it.

Workflow

1. Brief

Get the inputs above. If invoked from /repurpose, use the master Notion row's title + body as the brief.

2. Outline

Output a 5-8 line outline before drafting. Each line is one beat:

1. Hook        Opening sentence. The friction point.
2. Setup       Why this matters now. One paragraph of context.
3. Body 1      First insight / argument. Concrete example.
4. Body 2      Second insight. Counter-take or extension.
5. Body 3      Third insight. (Optional · drop if outline feels thin)
6. Resolution  What changes if you do this. Stakes.
7. CTA         The action. Specific.

Show the outline to the user. Wait for green light or revisions before writing the full draft. Don't draft the body until the outline is approved · revisions are expensive once the prose is flowing.

3. Draft

Once the outline lands, write the full piece.

Length targets:

TypeWordsBeatsNotes
Newsletter1200-15005-7Conversational. Personal. Time-anchored ("This week...").
Blog post800-15004-6More evergreen. SEO-aware subheads. Links out.
Substack issue1500-25006-8Longer arc. Can include footnotes / asides.

Structure constraints:

  • Lead with the hook in the first two sentences. Don't bury it.
  • Paragraphs are short (2-4 sentences each). White space is structural.
  • Subheads (H2) every ~300 words. Subheads are descriptive, not cute.
  • One specific concrete example per body section. Numbers, names, real situations · not generic.
  • The CTA is a single sentence. Don't tack on three.

Newsletter-specific:

  • Open with a date or time anchor ("Tuesday morning · 6 AM" · "Last week we shipped X").
  • Sign-off in your voice. No "Cheers," / "Best,". Just the next thought or a single line.
  • Subject line: separate output. See § "Subject lines" below.

Blog-specific:

  • Title is SEO-friendly but not slop. "Why GTM teams should stop building dashboards" beats "5 reasons your dashboard sucks".
  • Include 2-3 internal links (to other <YOUR_BRAND> blog posts or lead magnets) where they earn placement · never forced.

Substack-specific:

  • Allow more digressions and asides. Substack readers tolerate (and enjoy) tangential beats that wouldn't fit a blog.
  • Closing footnote with [1]-style reference if you cite anything.

4. Subject lines (newsletter only)

For newsletters, generate 3 subject-line options alongside the draft. Output them at the top of the deliverable. Patterns that work:

  • Specific number: "47 webinar signups, 0 demos"
  • Friction-first: "I almost killed our outbound stack last week"
  • Curiosity gap: "The one thing GTM Engineers get wrong"
  • One-word punch: "Outbound."
  • Question that reads like a thought: "What if your SDR didn't exist?"

Patterns to avoid:

  • Numbered listicles ("5 ways to...") · stale
  • Hyphenated compounds ("game-changing...") · slop
  • All caps · spam folder
  • Em dashes · slop
  • "Newsletter #042" · nobody opens that

5. Slop scrub (mandatory)

Run the brand-doc skill checklist on the final draft. The copy checks especially:

  1. Em-dash scan · zero — characters
  2. Banned-words scan · zero hits on leverage, synergy, robust, world-class, seamless, intuitive, powerful, innovative, delve, unleash, revolutionary, game-changing
  3. Negation-pivot scan · no "not just X, but Y" patterns
  4. Numerals · adjectives describing scale → real numbers where possible

If any check fails, fix and re-run. Don't ship.

6. Save the draft

Long-form drafts live in the Notion Content DB page body. Don't write the draft to a file in the repo (the repo is for assets, not text drafts).

Create a row via mcp__claude_ai_Notion__notion-create-pages with data source <YOUR_NOTION_CONTENT_DB_ID>:

  • Title: the piece's title (e.g. "Newsletter Issue 03 · The pipeline post")
  • Status: Scripting (until human reviews)
  • Pillar: pick the multi-select that matches the angle
  • Format: Long-form Article
  • Channel: Newsletter · Website (blog) · or as appropriate

Page body:

Type: <newsletter / blog / substack>
Length: <word count>
Source: <master URL if from /repurpose>

— Subject lines (newsletter only) —
1. <option 1>
2. <option 2>
3. <option 3>

— Outline —
1. <beat>
2. <beat>
...

— Draft —
<the full piece, with subheads>

If invoked from /repurpose, the orchestrator may handle row creation · check the source context to avoid duplication.

7. Report back

Hand off to the user:

  • Notion row URL
  • Word count
  • Subject lines (newsletter)
  • Confirmation all 4 copy checks passed

Suggest next: open the Notion row to review, then flip Status to Scheduled when ready, then post manually (you handle publishing).

Don'ts

  • Don't pad to hit length. If 1500 words feels thin, propose a shorter shape (LinkedIn post or thread) and stop. Padding produces the slop pattern you want to avoid.
  • Don't draft before the outline is approved. Rewrites in prose cost 5x more than rewrites in outline.
  • Don't impose newsletter time-anchors on blog posts. Blog evergreen voice means no "this week" openers.
  • Don't write listicles. "5 ways to..." structures are short-form posts, not long-form. If the topic naturally lists, write a thread or a LinkedIn post instead.
  • Don't forget the subject lines for newsletters. Three options. Always.
  • Don't include AI-flavoured intros ("In this piece, I'll explore how..."). Cut the ritual sentences and start with the hook.
  • Don't mix print and motion brand systems (this skill is text-only · brand-doc print rules apply if any visual treatment is added).
  • Don't fabricate quotes / metrics / customer names. If you're drawing from /repurpose source or a customer-interview transcript, use exact quotes only when the source is non-NDA · paraphrase otherwise.

Content graph (close the loop)

This skill writes to content/posts/, which feeds the content knowledge graph (content/CONTENT-WIKI.md). Before drafting, pull the relevant note from content/concepts/ and build on its canonical lines instead of starting blank. After the piece is saved, deposit its 1-3 core ideas back: add the post to each concept's ## Appears in (create a note if the idea is genuinely new), update any content/entities/ or content/mocs/ it touches, and append to content/log.md. Edges live in the concept notes (Obsidian backlinks are automatic), so don't edit the post body to add them.

See also

  • Brand reference: <YOUR_BRAND_DOC> · voice rules + skill checklist (if you have a brand doc with voice rules, link it here)
  • Voice corpus (shared with /linkedin-copywriter): ../linkedin-copywriter/corpus.md
  • Sibling skill (spoken long-form): /youtube-script
  • Notion Content DB data source: <YOUR_NOTION_CONTENT_DB_ID>
  • Upstream skill: /researcher (sources ideas), /repurpose (orchestrates from a master into long-form variants)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.