x-thread-generator
Skill1 starsChanged 46 days agoMITView on GitHub
Turn a topic, blog URL or transcript into a 5–15 post X (Twitter) thread, then ship it in a way the PostOnce X MCP supports (lead tweet plus replies to paste, one Premium long post, or standalone tweets over several days). Use when the user asks for a Twitter thread, an X thread, a thread generator, or to turn an article, video or notes into a thread.
In this repository5 other files it gives its agents.
- postonceSkill
- x-bio-generatorSkill
- x-content-calendarSkill
- x-tweet-formatterSkill
- x-tweet-generatorSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 11
Write a thread that people read to the end.
- Distinctive: line 27
Every post under 280 characters (count emoji as 2).
- Distinctive: line 40
Don't tell the user the thread was posted as a thread.
≈ “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
X (Twitter) thread generator
Write a thread that people read to the end. Be clear up front: the PostOnce MCP publishes single posts. It can't publish a thread as a reply chain, so every thread ends with a publishing choice (below).
Before writing
Get: the source (topic, blog URL, transcript or notes), the account and its voice, who the thread is for, and the one promise the thread makes. If the source is a URL you can read, pull the specific steps, numbers and examples from it. Never invent numbers, results or quotes.
Structure
| Post | Job |
| 1. Hook | The promise and why to keep reading: a result, a list count or a strong claim. Must work alone. |
| 2 | Context or proof: why you're worth listening to on this, in one or two lines. |
| 3 to N-1 | One point per post. Numbered if it's a list. Each post makes sense if read alone. |
| Last | The takeaway in one line, plus one ask: follow, read the full post, or reply. |
- 5–15 posts. Cut any post that repeats another.
- Every post under 280 characters (count emoji as 2). Aim for 150–260.
- Short lines, concrete examples, one screenshot or chart where it proves a point.
- No "1/" filler in the hook; numbering like "1/9" is optional and only if the user likes it.
- Links go in the last post. Hashtags: none, or one in the last post.
Publishing choices
Offer these, in this order, and let the user pick:
- Lead post now, rest by hand. Publish or schedule post 1 with
create_post. Give the user the remaining posts, numbered, to paste as replies in the X app right after it goes live. Useget_postto hand them the live URL. - One long post (X Premium only). If the account had X Premium when it was connected, collapse the thread into a single post of up to 4,000 characters: keep the hook as the first lines, turn each post into a short paragraph or list item. Text over 4,000 is cut off, so count it.
- Standalone tweets over several days. Rewrite each strong point so it stands alone without "as I said above", then schedule them as separate posts (for example one a day) with
publish_at. Good for evergreen material.
Don't tell the user the thread was posted as a thread. It wasn't.
Output
Return the thread as a numbered list with each post's character count, then the recommended publishing choice and why. On approval, follow the postonce skill: confirm the X account and times (with timezone) before calling create_post, then report each post's status with get_post.
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.