Skill-Details

write-thread-essay

Writes a narrow, voice-specific thread-essay format for insider audiences.

ÜbereinstimmungMöglichGeprüft für verfassen von essays
Quellejbrukh/skillsExterne Quelle
Gemeldete Installationen25Nur Popularitätssignal

Vor Nutzung prüfen

Die automatische Prüfung bewertet Relevanz, nicht Sicherheit oder Empfehlung. Lies vor der Nutzung die Quellanweisungen.

Gespeicherte Quellvorschau

SKILL.md

Dieser Auszug wurde bei der Prüfung gespeichert. Die externe Quelle enthält die vollständige und aktuelle Version.

---
name: write-thread-essay
description: Write punchy, conviction-driven thread essays in the Brukhman voice. Use when user says "write a thread essay", "write a post about", "draft a thread", or wants a conviction-driven insider-audience piece. First-person, insider-to-insider, with contrarian framing and enumerated arguments.
---

# Write Thread Essay

Write a punchy, conviction-driven essay in the style of a crypto-AI insider's long-form post. The voice is first-person, direct, assumes reader fluency, and reads like an expanded Twitter thread that grew into a proper argument.

## When to Use

- User invokes `/write-thread-essay`, asks to "write a thread essay", "write a post about...", or "draft a thread"
- User wants a conviction-driven, insider-audience piece for social or blog publication

## Voice & Tone

Voice: founder talking to other founders and degens on a public timeline. Authoritative through familiarity, not formality. The writer has skin in the game and isn't pretending otherwise. Reader should feel like they're getting a take from someone who's been in the room, not a summary from someone who read about it.

### Voice Anchor

Match this register:

> Decentralized AI training is coming to your OpenClaw hardware. Meta is closing Llama, Apple is shipping RDMA clusters, and your Mac is about to become a decentralized training node. I think a lot of people are going to get local GPU compute imminently, which has major implications for decentralized training.

> Decentralized training remains a contrarian thesis, though the goal posts have moved dramatically. Two years ago, it was commonly said to be impossible to do training without fast interconnects and high end devices. Today, we have done it across a number of models and communication bandwidth optimization paradigms.

Note what works: first person ("I think"), present tense conviction, insider shorthand (no need to explain what Llama is), stacking claims as facts, goal-post-shifting rhetoric that acknowledges skeptics then dismisses them. Dashes used freely. @ mentions are native. Technical specifics sit next to opinion without friction.

**Voice anti-anchor (do NOT sound like this):**

> *"The world of artificial intelligence is rapidly evolving, and decentralized networks represent an exciting new frontier. These innovative platforms are revolutionizing how we think about compute resources. Let's explore how this works and why it matters."*

Note what fails: vague ("rapidly evolving"), hype ("exciting," "revolutionizing"), throat-clearing ("Let's explore"), no specifics, no insider authority. Every sentence could appear in any AI article on any topic.

**Anchor protocol:** Before writing each beat, mentally re-read the voice anchor above. The anchor is not background context; it is the target register. If your next paragraph would not fit naturally between two paragraphs of the anchor excerpt, rewrite it until it would.

### Core Tone Attributes

- **First-person conviction.** "I think," "In my view," "I want to highlight this point." The writer has a position and states it directly. No hedging the core thesis.
- **Insider-to-insider.** No parenthetical glosses. No explaining what GPUs are or what tokenization means. If you don't know, this piece isn't for you. References to projects use @ handles naturally.
- **Stacked claims as openers.** The opening doesn't build tension slowly; it drops 2-3 bold claims in rapid succession, then spends the rest of the piece justifying them.
- **Contrarian framing.** Explicitly name the skeptic position, then show how it's already been overtaken by events. "The goal posts are now: but who will ever use decentralized training networks?"
- **Measured speculation.** Distinguish between what's happened ("Pluralis has already completed a 7.5B parameter model parallel run") and what's projected ("communities of AI agents are going to pre-train and tokenize their own models"). Use "I think" and "in my view" for speculation, plai
Vollständige Quelle auf GitHub lesen (öffnet externe Seite)
Kontext

Verwandte Arbeit