Produce X/Twitter thread in one of 7 thread types (Thread / Story Thread / Hot Take / How-To Thread / Listicle Thread / Case Study Thread / Daily Tweets batch). Specificity-first, open-curiosity-loop structure, final-tweet CTA. 2-3 threads per week.
Scanned 9/6/2026
Install to Claude Code
npx -y skills add Miosa-osa/canopy --skill write-x-thread --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Write X Thread?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/miosa-osa-write-x-thread)More formats (shields.io, HTML) on the badges page.
---
name: write-x-thread
description: Produce X/Twitter thread in one of 7 thread types (Thread / Story Thread / Hot Take / How-To Thread / Listicle Thread / Case Study Thread / Daily Tweets batch). Specificity-first, open-curiosity-loop structure, final-tweet CTA. 2-3 threads per week.
signal:
mode: linguistic
genre: x-thread
type: inform
format: markdown
structure: w-x-thread-7-type
receiver: X audience + algorithm
receiver_capacity: low-to-medium
department: marketing
agent_affinity: [marketing-head, twitter-writer]
required_compartments:
audience_intelligence_system: 50
offer_architecture: 30
copy_messaging: 30
upstream_dependency: content-calendar
execution_mode: interactive
tier: structured_ai
temperature_gate: warm
required_tools: [file_read, file_write]
required_mcps: [filesystem]
required_integrations:
social_posting: twitter-api
scheduling: buffer-api
analytics: twitter-analytics
credentials_required: [TWITTER_BEARER_TOKEN, TWITTER_API_KEY, TWITTER_API_SECRET, TWITTER_ACCESS_TOKEN, TWITTER_ACCESS_SECRET]
evidence_gate:
- thread_type_selected
- hook_tweet_specificity_gte_8
- tweet_count_calibrated
- final_cta_tweet_present
- char_limits_respected
- voice_matched
- signal_score_gte_0_8
keywords:
- twitter thread
- x thread
- write thread
- twitter post
priority: 1
version: 1.0
---
# /write-x-thread — X/Twitter Thread (7 Types)
## Role
Produce X/Twitter threads in 7 canonical types. Lineage: Marketing department (7 thread types), direct-response X operators, content-to-conversion bridge applied to async-skim platform.
## The 7 Thread Types
| # | Type | Use |
|---|---|---|
| 1 | **Thread (standard)** | General teaching / analysis |
| 2 | **Story Thread** | Narrative with tension + reveal |
| 3 | **Hot Take** | Contrarian positioning, polarizing |
| 4 | **How-To Thread** | Step-by-step actionable |
| 5 | **Listicle Thread** | Ranked/numbered items |
| 6 | **Case Study Thread** | Real client result walkthrough |
| 7 | **Daily Tweets (batch)** | 5-7 single tweets for the week |
## Structure
```
Tweet 1 (HOOK) — 280 chars max, 120-200 ideal, earns the open
Tweets 2-N (BODY) — each tweet = 1 idea, paragraph-style, save-optimized
Final tweet (CTA) — single action
Optional: reply-to-thread — link / lead magnet / offer mention
```
Tweet count calibrated by type:
- Thread / How-To: 7-12 tweets
- Story Thread: 10-15 tweets
- Hot Take: 3-5 tweets
- Listicle: 6-11 tweets (5-10 items + hook + CTA)
- Case Study: 8-14 tweets
- Daily Tweets: 5-7 standalone tweets (batch output, not a thread)
## Decision Logic
### Hook Tweet Invariants
- First line earns the click (first 50 chars visible on some clients)
- Specificity ≥ 8
- Number or named-entity or bold claim
- Opens a loop or promises specific value
- NEVER "A thread 🧵" as the whole hook — that's lazy
### Engagement Patterns (X-specific)
- Replies > likes (signal value)
- Bookmarks > replies (highest value — means they'll return)
- Threaded replies are decay-prone — main thread matters most
- Quote-tweets expand reach
### Format Choices
- **Image in hook tweet:** 15-30% reach boost for some accounts (test)
- **Video in hook tweet:** higher engagement, longer dwell
- **Text-only:** still works for authority/hot takes
- **Poll hook tweet:** engagement-heavy, less save-ability
### Voice-Match
- 3+ `phrases_to_use` across thread
- Creator's sentence rhythm preserved
- Banned vocabulary swept
### The Final-Tweet CTA
Single action:
- "Follow for more"
- "DM me for the playbook"
- "Link to full breakdown" (in reply, not in thread body)
- "Book a call" (for offer-tied threads only — 20% rule)
## Tacit Principles
1. **First 50 chars visible on many clients.** Hook must open in that window.
2. **Save-optimize** — threads that get bookmarked get re-surfaced. Structured lists + frameworks + checklists.
3. **One idea per tweet.** Don't cram. Threadability = readability.
4. **Polarize for hot takes.** Mild takes get zero traction.
5. **Story threads need tension.** Arrival → Adversity → Aftermath (Jenny Hoyos 3-A).
6. **No external link in thread body.** Link in last tweet or in reply.
7. **Quote-tweet your own thread** 24-48h later with a takeaway — 2× reach.
8. **Batch daily tweets on one topic.** 5 tweets around one theme > 5 random tweets.
9. **Reply to commenters** within 1-2 hours of posting.
10. **Don't over-thread.** 15+ tweets = drop-off steep. 7-12 is the sweet spot.
## Process
### Phase 0 — Load
- Calendar brief, compartments 1-5, voice doc
- `reference/frameworks/primitives/specificity.md`
### Phase 1 — Thread Type Selection
Per calendar brief pillar + stage. Or `--type=1-7`.
### Phase 2 — Hook Tweet
Draft 3-5 variants. Specificity ≥ 8. Select primary.
### Phase 3 — Body Draft
Per type:
- **Thread:** 1 idea per tweet, 7-12 total, logical progression
- **Story:** 3-A structure spread across 10-15 tweets
- **Hot Take:** 3-5 tweets — premise + support + sharpest point
- **How-To:** numbered steps, 7-12 tweets
- **Listicle:** countdown or ascending numbered list, 5-10 items
- **Case Study:** specific outcome → context → journey → mechanism → takeaway
- **Daily Tweets:** 5-7 standalone tweets on one theme (batch)
### Phase 4 — Final CTA Tweet
Single action. No stack of asks.
### Phase 5 — Voice-Match + Compliance
- 3+ phrases_to_use
- Banned vocab swept
- No income guarantees
- Specificity check per tweet
### Phase 6 — Visual Decisions
- Image/video in hook tweet (optional but tested)
- GIFs sparingly (not in every tweet)
### Phase 7 — Publishing Plan
- Optimal times: 9-11am + 6-8pm ET (adjust per audience tz)
- Reply-to-thread 24-48h later for lift
- Quote-tweet with takeaway 48-72h later
## Output Format
```markdown
# X Thread — [Creator Brand] — [Topic]
**Calendar Brief ID:** [e.g. W1-TW-1]
**Thread Type:** [1 Thread | 2 Story | 3 Hot Take | 4 How-To | 5 Listicle | 6 Case Study | 7 Daily Tweets]
**Tweet Count:** [N]
**Hook Specificity:** [N/10]
**Publish Time:** [window]
**Version:** 1.0
---
## Thread
### Tweet 1 — HOOK (⬇️ indicator included)
```
[hook copy — 120-200 chars, visible on first line, ends with "🧵" or "⬇️" or number indicator]
```
### Tweet 2
```
[one idea, 180-280 chars]
```
### Tweet 3
```
[...]
```
...
### Final Tweet — CTA
```
[single action — "Follow for more" | "DM me 'X'" | "Full breakdown: [link]"]
```
---
## Reply-to-Thread (scheduled T+24-48h)
```
[takeaway tweet — quotes own thread, drops 1 more insight, re-surfaces]
```
## Quote-Tweet Plan (T+48-72h)
```
[quote own hook with takeaway]
```
---
## Hook Variants (not selected)
1. [variant 1]
2. [variant 2]
## Voice-Match Audit
- phrases_to_use placements: [count]
- Banned vocabulary: [clean]
## Compliance
- [ ] No income guarantees
- [ ] Specificity ≥ 8 on hook
- [ ] FTC disclosure if testimonials
## Publishing Plan
- Publish time: [date + window]
- Reply-to-thread: T+24-48h
- Quote-tweet: T+48-72h
- Engagement block: T+0 to T+90 min
## Repurpose Chain
- Derivative 1: /write-linkedin-post
- Derivative 2: /write-reel (thread main insight as short)
- Derivative 3: /email-sequence (broadcast referencing thread)
---
## Appendix A — Thread structure diagram
[visual of tweet-by-tweet narrative flow]
## Appendix B — Per-tweet save-ability audit
[which tweets most bookmark-able]
```
## Important Rules
- **NEVER exceed 15 tweets in a single thread** (steep drop-off).
- **NEVER "A thread 🧵" as the whole hook.**
- **NEVER external link in thread body** — final tweet or reply.
- **NEVER stack multiple CTAs.**
- **ALWAYS hit specificity ≥ 8 on hook.**
- **ALWAYS 1 idea per tweet.**
- **ALWAYS reply-to-thread 24-48h for lift.**
- **ALWAYS voice-match.**
## Verification Checklist
- [ ] Thread type selected
- [ ] Hook specificity ≥ 8
- [ ] Tweet count calibrated to type
- [ ] 1 idea per tweet
- [ ] Character limits respected (280 max, 180-280 ideal)
- [ ] Final CTA tweet — single action
- [ ] Reply-to-thread scheduled
- [ ] Quote-tweet plan scheduled
- [ ] Voice-matched (3+ phrases_to_use)
- [ ] Compliance passed
- [ ] Triple-layer S/N ≥ 0.8
## Next Skills
- `/write-linkedin-post` — derivative
- `/write-reel` — short-form version
- `/email-sequence` — broadcast reference
## References
- Marketing department Department (7 thread types)
- `reference/frameworks/primitives/specificity.md`
- `reference/canonical/spec/INTEGRATIONS.md` (Twitter API)
---
*Version 1.0 — 2026-04-19. Cycle 3 Attention — X/Twitter.*
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!