tweet-thread-builder
Skill1 starsChanged 38 days agoMITView on GitHub
Turn one idea into a tight tweet thread - a hook that earns the click, one beat per post, and momentum to a single CTA. Use when someone asks "turn this into a thread", "write a thread about my launch", "why did my thread flop", or "give me hook options for this". Do NOT use for LinkedIn-format posts - use linkedin-post-writer instead. Do NOT use for adapting one finished piece across many platforms - use content-repurposing instead.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- SkillMedev/social-media-studioword for word, first seen Oct 5, 2026
Other versions1 project uses a version of this.
Word for word · 1
- SkillMedev/social-media-studio★ 1100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- brand-guidelinesSkill
- case-study-builderSkill
- content-briefSkill
- email-newsletter-proSkill
- landing-page-copySkill
- linkedin-post-writerSkill
- testimonial-capture-interviewSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
No preamble. The first line is the whole pitch.
- Distinctive: line 29
Always draft two hooks with different mechanisms so the user can pick or test:
- Distinctive: line 55
Read the thread top to bottom once in one sitting.
- Distinctive: line 93
Do not open with "A thread 🧵" and no substance - the hook must contain the promise, not announce the format.
- Distinctive: line 94
Do not exceed one idea per post.
- Distinctive: line 95
Do not stack CTAs (follow + link + reply) - each added ask cuts response to all of them.
- Distinctive: line 96
Do not pad to hit a post count.
- Distinctive: line 97
Do not write abstractions where the user has numbers - "grew a lot" when the material says "34%" is malpractice.
- Distinctive: line 103
Every post's last line creates pull toward the next.
≈ “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.
Files
Tweet Thread Builder
Threads live or die on the first post: 90%+ of potential readers see only the hook, so a mediocre hook wastes every good post behind it. The costly mistake is spreading one thin idea across fifteen posts or burying the payoff behind "a thread 🧵" with no substance. Earn each scroll.
Operating procedure
Step 1: Gather inputs
- The single idea or story - one sentence. If the user gives several, pick the one with the most concrete specifics (numbers, names, before/after); park the rest.
- The payoff for the reader: what they know or can do after reading.
- Proof material: numbers, screenshots, examples. A thread without at least two concrete specifics is an opinion, not a thread - send the user back for material.
- The one CTA (default: follow; alternatives: a link, or "reply with your version").
Label any number the user estimates as an estimate.
Step 2: Write the hook (post 1) - and an A/B pair
- Promise a specific payoff: a number, a transformation, a contrarian take.
- No preamble. The first line is the whole pitch. Keep the hook under 200 characters - dense hooks read as work.
- Proven patterns: "I did X for N days. Here's what happened." / "Everyone believes X. They're wrong." / "N lessons from [hard thing]:"
Always draft two hooks with different mechanisms so the user can pick or test:
- A (curiosity/story): "I deleted 80% of our landing page. Signups went up 34%. What we removed:"
- B (contrarian/claim): "Long landing pages are killing your signups. We cut ours to one screen and grew 34%. Proof:"
Same material, different bet: A sells the mystery of what was removed; B sells the argument. Pick A for story-driven audiences, B when the take itself is polarizing enough to drive replies.
Step 3: Outline the beats
- One idea per post. If a post has two ideas, split it. If two posts share one idea, merge them.
- 5-12 posts is the sweet spot. Under 5, make it a single tweet; over 12, cut - every post past 12 sheds readers faster than it adds value.
- Order the beats so each ends on a small open loop that the next post resolves; momentum is the structure.
Step 4: Write the body posts
- Make it concrete: examples, numbers, and specifics over abstractions.
- Short lines and white space; threads are read on phones.
- Front-load each post - the first line of every post is a mini-hook.
Step 5: Write the close
- Restate the single takeaway in one post.
- One CTA only. Optional: a one-line summary post people can quote-tweet.
Step 6: The scroll test
Read the thread top to bottom once in one sitting. At any post where you would stop scrolling, fix that post or cut it. Then check the numeric guardrails: 5-12 posts, hook under 200 characters, at least two concrete specifics across the body, exactly one CTA.
Worked example thread (6 posts)
1/ I sent 200 cold emails with a 2% reply rate.
Rewrote one sentence. Reply rate hit 11%.
The sentence, and why it worked:
2/ The old opener: "Hi, I'm the founder of [company], we help teams do X."
Nobody cares who you are in sentence one.
They care whether you know THEM.
3/ The new opener: "Saw you're hiring 3 SDRs - usually means outbound
is working but reporting is a mess."
One observed fact + one inferred pain. That's it.
4/ Why it works: it proves 30 seconds of research.
2% of cold email gets replies because 98% of it could be
sent to anyone. Specificity is the whole game.
5/ The formula: [thing you observed about them] + [what that
usually means] + one-sentence ask.
Under 60 words total. No deck attached.
6/ Takeaway: your first sentence should be about them, not you.
Steal the formula, run 50 sends, check your reply rate.
Follow @[handle] - I share one outbound experiment a week.Why it passes: 6 posts (in range), hook is 118 characters with two numbers and an open loop, every post carries exactly one idea, posts 1-5 each end pulling toward the next, and the close restates the takeaway with one CTA.
Deliverable
Produce a numbered, post-by-post thread draft containing: two hook variants labeled A and B, 5-12 body posts each carrying one idea, a closing post with the takeaway and one CTA, and a one-line note on which hook to prefer and why.
Do NOT
- Do not open with "A thread 🧵" and no substance - the hook must contain the promise, not announce the format.
- Do not exceed one idea per post; splitting is free, cramming costs readers.
- Do not stack CTAs (follow + link + reply) - each added ask cuts response to all of them.
- Do not pad to hit a post count; a tight 6 beats a padded 12 every time.
- Do not write abstractions where the user has numbers - "grew a lot" when the material says "34%" is malpractice.
Quality bar
- Hook under 200 characters, contains a specific payoff, and would stop your own scroll.
- 5-12 posts, one idea each, at least two concrete specifics in the body.
- Every post's last line creates pull toward the next.
- Exactly one CTA; the takeaway is stated, not implied.
- The thread survives the top-to-bottom scroll test with zero stop-points.
To adapt the finished thread into a LinkedIn post, newsletter section, or other formats, route to content-repurposing; for a native LinkedIn version written from scratch, use linkedin-post-writer.
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
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.

