SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsx-short-content-craft

x-short-content-craft

Marketing & growth

View repository

Updated 10-15 23:44

About this skill

Short-Form Content Types and Structures — Four Types + Hook-Body-CTA + Viral Formula

R — Original Text (Reading)

Four types of short-form content: Story type (Hook-Body-CTA, emotional resonance), List type (structured organization, easy to bookmark), Opinion type (unique perspective + controversy, viral spread), Tutorial type (Step-by-Step, building authority). Shared structure: 1. The Hook: grab attention within 3 seconds in the first line; 2. The Meat: break down into short single sentences or bullet points; 3. The CTA: guide interaction at the end (RT/Reply/Follow). Viral formula: one genuinely useful tool + one clear scenario + a usage path within three steps.

— AIP Going Global Tutorial 4.1/4.6; Xiangyang Qiaomu, X Viral Secrets Sharing · Viral Content Formula

I — Methodology Skeleton (Interpretation)

X short-form content falls into four types, each serving a different metric — choosing the wrong type equals a goal-structure mismatch:

  • Story type → drives follower growth (emotional resonance makes people want to follow you)
  • List type → drives bookmarks (structured information is easy to save and easy to share)
  • Opinion type → drives interaction (controversy sparks replies and reposts)
  • Tutorial type → drives loyalty (Step-by-Step builds trust and authority)

The four types share a Hook-Body-CTA three-part skeleton:

  • Hook: grab attention within 3 seconds in the first line (counterintuitive fact / strong opinion / How-to opening / numerical suspense)
  • Body: break long paragraphs into short single sentences or bullet points, leave blank lines between paragraphs, and make good use of emoji for visual guidance
  • CTA: guide interaction at the end (RT / Reply / Follow / comment a keyword to get resources)

Resource/tool-type content can additionally apply theviral formulafor quick drafting:

Viral = genuinely useful tool + clear usage scenario + usage path within three steps

Each of the three elements plays its role: tool = the core of value, scenario = "when would I use this", steps = "can I start right now". "Within three steps" forces creators to lower the reader's cost of action.

A1 — Application in the Book (Past Application)

Case 1: Vivek Burman — Story-type post (74K views / 567 likes / 191 bookmarks)

  • Problem: wants to share a personal experience of getting high exposure with a short video and attract creators' attention
  • Use of the methodology: chose the story type. The Hook uses numerical suspense — "Yesterday, I posted 1 short, and within the first 5 hours, it peaked to 314,000 views per hour. How?"; the Body shares the CTA strategy ("Make this go viral so the creator can react") and personal experiment results; the CTA uses "Follow + Like + Comment 'CTA' And I will send you my top 5 CTA" to trigger interaction in exchange for value
  • Conclusion: the story-type Hook-Body-CTA drives follower growth — the hook targets the pain point + hints at the result, the body unfolds the personal experience, and the CTA guides action
  • Result: 74K views, 567 likes, 191 bookmarks (likes > bookmarks, matching the metric characteristics of the story type driving follows rather than bookmarks)

Case 2: Alex Finn — List-type post (22K views / 355 likes / 297 bookmarks)

  • Problem: Want to share how to use the X platform's List tool
  • Use of the methodology: Choose the list type. Hook = "Lists are the most powerful tool on X"; the Body is structured and expanded into three lists (Interact / Networking / Inspiration), with each list further broken down into uses and criteria for adding; the ending provides a "How to USE THESE LISTS" application guide
  • Conclusion: List-type structured organization, with numbering + bullet points for easy scanning and saving
  • Results: 22K views, 355 likes, 297 saves (save rate 83%, far exceeding likes, validating "list type drives saves")

Case 3: Aaron — tutorial-type post (45K views / 314 likes / 410 saves)

  • Problem: Wants to teach people how to become an authority in a field
  • Use of the methodology: Choose the tutorial type. Hook = "How to become an authority in any space"; the Body uses bullet points to unfold step by step (Start an X account → Split content 50/50 → Repeat all for as long as you possibly can); focuses on replicable methods
  • Conclusion: The tutorial-type Step-by-Step builds trust, saves > likes validating the loyalty metric
  • Results: 45K views, 314 likes, 410 saves (saves > likes, matching the tutorial type's trait of driving loyalty)

A2 — Trigger Scenario (Future Trigger) ★

In what situations would users need this skill?

  1. The user wants to draft a tweet/post but doesn't know what structure or type to use
  2. The user asks "how to post a post teaching people to use XX tool" → tutorial type + viral formula
  3. The user asks "how to write a tweet that's easy to get saved" → list type
  4. The user has a controversial opinion and asks how to format it to spark more engagement → opinion type
  5. The user has a resource/tool to share and asks how to write it → apply the viral formula (tool + scenario + three steps)

Language signals (activate when these appear in the user's words)

  • "怎么写推文/帖子" / "how to write a tweet/post"
  • "用什么结构/排版" / "what structure / how to format"
  • "教人用 XX" / "tutorial post / how-to"
  • "怎么发一条 XX 的帖" / "how to post about XX"
  • "Hook / 钩子" / "CTA / call to action"
  • "怎么让人收藏/互动/关注" / "how to get saves/replies/follows"

Distinction from adjacent skills

  • Difference from x-content-archetypes : archetypes divides into four proto-types by "what the reader can do after receiving it" (resource gateway/tool tutorial/AI tool discovery/ordinary expression), with the purpose ofpredicting the spread ceilingandauditing content structure; this skill divides into four types bynarrative structure(story/list/opinion/tutorial), with the purpose ofchoosing a skeleton for drafting. A post can simultaneously be "resource gateway type" (archetype) + "list type" (craft); the two classification sets are orthogonal.
  • Difference from x-five-piece-checklist : checklist is pre-publicationchecking the completeness of the five-piece set(value promise/scenario/gateway/evidence/reason to save); this skill is thedrafting stageof choosing a structure. Draft first (this skill) → then check (checklist).
  • Difference from x-three-translations : translations is about turning announcement-style internal languageinto external language that readers can take away(publish→help, capability→scenario, conclusion→evidence); this skill is the selectionstructural skeleton. First translate the language (translations) → then apply the structure (this skill).

E — Executable Steps (Execution)

When the skill is activated, the agent should execute the following steps:

  1. Identify the target metric, choose the content type

    • Want to gain followers → story type; want to be bookmarked → list type; want to spark interaction → opinion type; want to build loyalty → tutorial type
    • If the user has not specified a goal, ask "What result do you most want from this post: followers, bookmarks, interaction, or loyalty?"
    • Completion criteria: Clearly state which type was chosen and the corresponding target metric

  2. Draft according to the Hook-Body-CTA skeleton

    • Hook: Write a first line that grabs attention within 3 seconds (counterintuitive fact / strong opinion / How-to opening / numerical suspense)
    • Body: Break the content into short single sentences or bullet points, leave blank lines between paragraphs, use emoji for visual guidance but do not overuse
    • CTA: End with a call to interaction (RT / Reply / Follow / comment a keyword to get resources)
    • Completion criteria: All three sections are present, with at least one sentence each for Hook / Body / CTA

  3. If it is resource/tool-type content → apply the viral formula check

    • Viral = genuinely useful tool + clear use scenario + usage path within three steps
    • Stop condition: If it is not resource/tool-type (such as pure opinion/pure story), skip this step
    • Completion criteria: All three elements are present (tool name/scenario description/≤3-step path), or confirm it is not applicable

  4. Self-check type-metric match

    • Story type: Does it have emotional narrative and a story arc?
    • List type: Is it structured with numbering/bullet points?
    • Opinion type: Does it have a unique perspective and controversy? Keep it concise and not neutral?
    • Tutorial type: Does it have a Step-by-Step replicable process?
    • Completion criteria: The structural features of the chosen type are reflected and consistent with the target metric

B — Boundary (Boundary) ★

Do not use this skill in the following situations

  • Already have a draft and want to check completeness → use x-five-piece-checklist(five-piece set: value promise/scenario/entry/evidence/reason to bookmark)
  • Want to rewrite announcement-style language into help-style language → use x-three-translations(publish→help, capability→scenario, conclusion→evidence)
  • Want to classify by "reader function" to predict the spread ceiling → use x-content-archetypes(resource entry/tool tutorial/AI discovery/ordinary expression)
  • Pure information query(such as "What is the character limit for X posts?") → answer directly, no drafting framework needed

Failure modes the author warns about in the book

  • ce12 long-form superstition: Writing long without choosing a type. X is a short-content platform; long does not equal traffic; first choose the type to define the structure, then refine, rather than piling on words. Long-form reduces the probability of "first-glance entry," and the X editor has many formatting issues.
  • ce15 overfitting data as a formula: The golden length of 120-220 characters and similar claims are overfitting results by Xiangyang Qiaomu based on personal 3.4G data, and the author himself labels them "for reference only." Word count is a reference, not a hard metric; choosing the right type + Hook-Body-CTA is more important than padding the word count.

The author's blind spots / limitations of the era

  • The four-type × metric mapping (story → follower growth/list → saves/opinion → engagement/tutorial → loyalty) is an empirical generalization, not causal proof. All 4 case posts are samples of successes, which may reflect timing/luck rather than the causal effect of the type itself.
  • The viral formula comes from overfitted samples in a specific niche (AI/product R&D/Chinese-speaking audience, 2023-2026 algorithm versions). Data distributions differ across niches/audiences/periods, so applying it directly may not produce the expected results.

Easily confused adjacent methodologies

  • x-content-archetypes It also discusses the "four types," but categorizes byreader function(resource entry/tool tutorial/AI discovery/ordinary expression), with the goal of predicting spread; this skill categorizes bynarrative structure(story/list/opinion/tutorial), with the goal of guiding drafting. The two classification systems are orthogonal—the same post has a place in each system.

Related skills

  • depends-on x-content-archetypes: Before drafting, first understand the spread differences among the four archetypes; the archetype determines which short content type to use—classify and predict first, then choose the skeleton for drafting.
  • composes-with x-five-piece-checklist: After drafting, use the five-piece set to check completeness; one creates and the other checks—produce first, then find gaps.

Audit information

  • Validation passed: V1 ✓ / V2 ✓ / V3 ✓
  • Test pass rate: 100% (fallback self-assessment, independent blind testing recommended) (see test-prompts.json)
  • Distillation time: 2026-07-14