agentleFS
jsathjsath/claude-skills/skills/writing/newsletter/SKILL.md

newsletter

Skill14 starsChanged 35 days agoMITView on GitHub

Writes a personal founder newsletter issue in the user's own voice (learned from 3 to 5 of their past issues): a real firsthand scene first, one lesson, one thing readers can use today, one call to action. Also generates a deduplicated, ranked idea bank and, optionally, audits past issues' open and click data correctly. Use when someone asks to write this week's newsletter, turn a story or brain dump into an issue, fix a draft that sounds like AI, or come up with newsletter ideas. Works with any email platform; drafts only, never sends or schedules.

23 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use lead-magnet, linkedin and 5 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • robertbstillwellSkill

    lead-magnet

    Create lead magnets and gated content to capture email subscribers and leads. Use when asked to design checklists, templates, calculators, mini-courses, whitepapers, or other downloadable content offers. Trigger phrases: "lead magnet", "gated content", "email capture", "content offer", "free download", "checklist", "template", "whitepaper", "swipe file", "content upgrade", "opt-in offer".

    Also in 4 projects that use this

  • kkoppenhaverSkill

    linkedin

    Write LinkedIn posts in Keanan's voice. Use when drafting LinkedIn content, creating professional posts, or sharing builder-focused insights. Triggers include requests to "write a LinkedIn post", "draft something for LinkedIn", "help me post on LinkedIn", or any LinkedIn content creation. Generates upbeat, builder-centric content mixing practical how-tos with candid humor.

    Also in 3 projects that use this

  • Dokkabei97Skill

    brand-voice

    Defines and enforces brand voice guidelines for startups. Creates tone, style, messaging frameworks and provides consistency checks across all content. Ensures every piece of communication sounds like it comes from the same brand.

    Also in 4 projects that use this

  • silvainfmSkill

    referral-program

    When the user wants to create, optimize, or analyze a referral program, affiliate program, or word-of-mouth strategy. Also use when the user mentions 'referral,' 'affiliate,' 'ambassador,' 'word of mouth,' 'viral loop,' 'refer a friend,' or 'partner program.' This skill covers program design, incentive structure, and growth optimization.

    Also in 3 projects that use this

  • silvainfmSkill

    launch-strategy

    When the user wants to plan a product launch, feature announcement, or release strategy. Also use when the user mentions 'launch,' 'Product Hunt,' 'feature release,' 'announcement,' 'go-to-market,' 'beta launch,' 'early access,' 'waitlist,' or 'product update.' This skill covers phased launches, channel strategy, and ongoing launch momentum.

    Also in 3 projects that use this

  • indranilbanerjeeSkill

    keyword-research

    Research keywords via the connected data source: intent, gaps. Grouping → keyword-cluster. \"what keywords should we target\"

    Also in 3 projects that use this

  • +1 more from silvainfm

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Ask for 3 to 5 past issues the user is proud of (paste, files, or the public archive URL).

    line 14
  2. Distinctive:

    Substitution test: if another founder could send the body unchanged except for city, company and tool names, it is not personal enough.

    line 47
  3. Distinctive:

    Prefer their plain question ("how do you add AI to your business when it changes daily?") over consultant language ("navigating an evolving …

    line 49
  4. Distinctive:

    Never invent feelings, self-deprecation or "lol" to sound casual.

    line 50
  5. Distinctive:

    Never: em dashes, "here's the thing", "let that sink in", "unlock", "game changer", "deep dive", "level up", "→ try it:" after every …

    line 51
  6. Distinctive:

    No private finances, client names or deal mechanics unless the user confirms they are public.

    line 52
  7. Distinctive:

    When the user says one is a duplicate, replace it in the same reply.

    line 80
  8. Distinctive:

    Prefer an honest experiment ("here's the setup and the scorecard, results next issue") over invented success.

    line 82
  9. Very specific:

    Only when asked "what's working" and the user can export per-issue stats (any platform's CSV or API).

    line 86

“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “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

Newsletter

A personal email that happens to contain value, not a blog post or a framework dump.

By Johann (@johannships). Built from editing my own weekly issues and auditing every sent issue's opens and clicks.

Step 0: learn the voice (first run, then reuse)

Ask for 3 to 5 past issues the user is proud of (paste, files, or the public archive URL). Write voice.md next to this file:

  • greeting and sign-off (copy them exactly; they repeat every issue)
  • capitalisation, paragraph length, header style
  • words and shorthand they use; words they never use
  • how they move from story to lesson (quote 2 real transition lines)
  • CTA style and where it sits

Show it once, apply corrections, reuse after. No samples: use the defaults below and label the draft "default voice".

Inputs

  • Raw material (required): a brain dump, voice-note transcript, a story, or "what happened this week". If you only have a topic or an itinerary, stop and ask for one short messy note: what happened, what surprised you, what you got wrong. Ask before writing a full draft, not after it is rejected.
  • The one job of this issue: replies, a community join, a booked call, a download. One primary CTA. Ask if not given.
  • Links for the CTA and any resource. Keep missing ones as literal placeholders like [COMMUNITY LINK]. Never invent a URL.
  • Optional: the archive URL, for dedupe.

Structure

  1. Subject: lowercase-friendly, 3 to 7 words, sells the story ("i left the country. my AI ran my business."). Not "How I Achieved Success Through Strategic Planning".
  2. Preview line: promises the concrete payoff ("hint: it's not about the money").
  3. Greeting + scene: open inside a real moment. What you were doing, what you saw, what surprised you.
  4. One lesson the story earns. 2 to 3 sentences.
  5. The useful part: a method, checklist, prompt or exercise they can use today, written conversationally. Bullets are fine; a five-part numbered framework usually reads like a content marketer.
  6. What you're building / where you're headed: 2 to 3 short paragraphs, optional.
  7. One CTA, after the value. A free resource is value, not a CTA: give it before the ask.
  8. Sign-off: same structure every issue (from voice.md).

[image placeholder: what photo] only where a real photo would help.

Paragraphs 1 to 3 sentences, often 1. Lots of whitespace. Short ### headers.

Voice rules

  • Substitution test: if another founder could send the body unchanged except for city, company and tool names, it is not personal enough. Get a real detail from the user.
  • Conservation mode: when the user hands you rough copy and says it is more in their voice, keep their words, claim order, fragments, asides and closing thesis. Fix typos, grammar, repetition and transitions only. No polished new opener, no tidy moral, no expansion after they asked for shorter.
  • Prefer their plain question ("how do you add AI to your business when it changes daily?") over consultant language ("navigating an evolving landscape").
  • Never invent feelings, self-deprecation or "lol" to sound casual.
  • Never: em dashes, "here's the thing", "let that sink in", "unlock", "game changer", "deep dive", "level up", "→ try it:" after every section.
  • No private finances, client names or deal mechanics unless the user confirms they are public.

Example (default voice, excerpt)

subject: i stopped building software
preview: not because it doesn't work

hey friends,

i stopped building saas.

not because it doesn't work. it does.

[scene: the moment you decided]

### the one thing that made the difference

[lesson in 2 to 3 sentences]

paste that into Claude. say "build me a system that handles this." you'll be surprised how fast you have a working version.

Mode: ideas

  1. Dedupe first: read the public archive (all pages) and any drafts the user shares. Compare scene, lesson, asset and CTA, not just titles. If the archive is unreachable, say so.
  2. Return the number asked for (default 10), genuinely distinct. For each: subject, preview, opening scene, reader takeaway, practical asset, the one CTA, one sentence on why it should convert.
  3. Rank them and name the strongest overall and the strongest for the issue's job.
  4. When the user says one is a duplicate, replace it in the same reply.

Prefer an honest experiment ("here's the setup and the scorecard, results next issue") over invented success.

Mode: performance audit (optional)

Only when asked "what's working" and the user can export per-issue stats (any platform's CSV or API). Follow references/engagement-audit.md. Read-only: never create, schedule or send anything in the platform.

Output

The full issue in one fenced markdown block (subject, preview, body, CTA, sign-off) so formatting survives a paste. Commentary outside the block, 3 lines max. If a claim needs checking, put internal note (remove before sending): ... above the block.

Checklist before handing over

  • Real scene comes before any framework
  • Passes the substitution test
  • Exactly one primary CTA; resources come before it
  • Every number traces to the user's material
  • No em dashes, no banned phrases (search the text, do not eyeball)
  • Subject and preview each do a different job
  • Would the user actually send this? If it sounds generated, rewrite.

Feedback log

Append corrections and issue results to feedback.md (date, issue, what changed, rule going forward). Read it first next time.

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.