Communitygithub.com

aaione/everything-claude-code-zh

Multi-platform content distribution across X, LinkedIn, Threads, and Bluesky. Uses content-engine mode to adapt content per platform. Never posts identical content across platforms. Ideal for users who want to distribute content across social platforms.

What is everything-claude-code-zh?

This AI agent skill enables seamless multi-platform content distribution across X, LinkedIn, Threads, and Bluesky, ensuring each platform receives a tailored version of the core idea. It emphasizes platform-specific adaptation over copy-paste posting, preserving the author's authentic voice while respecting each platform's constraints. The skill guides users through a structured workflow: starting from the strongest source material, capturing the brand voice via the `brand-voice` skill, then adapting content for each platform's unique tone and formatting rules (e.g., compressed for X, contextual for LinkedIn, direct for Threads, concise for Bluesky). It enforces strict quality checks to avoid generic filler, forced professionalism, or cross-platform clones, and integrates with companion skills like `brand-voice`, `content-engine`, and `x-api` for a complete publishing pipeline. This is particularly useful for developers, technical writers, and creators who need to maintain consistent messaging across diverse audiences without sacrificing platform-specific effectiveness.

Works with✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/aaione/everything-claude-code-zh/tree/HEAD/skills/crosspost

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

Cross-Platform Publishing

Distribute content across multiple platforms without turning it into the same fake post wearing four different outfits.

When to Activate

  • User wants to publish the same core idea on multiple platforms
  • A single post, update, version, or article requires platform-specific adaptations
  • User says "cross-post this", "share this everywhere", or "adapt this for X and LinkedIn"

Core Rules

  1. Do not post the same copy across platforms.
  2. Preserve the author's voice on every platform.
  3. Adapt for constraints, not stereotypes.
  4. Each post should still be about one thing.
  5. If the original material doesn't earn a CTA, question, or moral, don't invent one.

Workflow

Step 1: Start with the Primary Version

First, choose the strongest source version:

  • Original X post
  • Original article
  • Release notes
  • Tweet thread
  • Memo or changelog

If the source material still needs voice shaping, first use content-engine.

Step 2: Capture Voice Fingerprint

If the source voice hasn't been captured in the current session, first run brand-voice.

Reuse the generated VOICE PROFILE directly. Do not build a second temporary voice checklist here unless the user explicitly wants a fresh coverage for this marketing campaign.

Step 3: Adapt by Platform Constraints

X

  • Keep it compressed
  • Open with the sharpest opinion or work
  • Use tweet threads only when a single post can't carry the argument
  • Avoid hashtags and generic filler

LinkedIn

  • Only add background that non-domain people need
  • Don't turn it into a fake founder reflection post
  • Don't add a closing question just because it's LinkedIn
  • Don't force a "professional tone" if the author's original style is sharper

Threads

  • Keep it readable and direct
  • Don't write fake ultra-casual creator copy
  • Don't paste the LinkedIn version and then shorten it

Bluesky

  • Keep it concise
  • Preserve the author's rhythm
  • Don't rely on hashtags or feed-manipulation language

Publishing Order

Default:

  1. Post the strongest native version first
  2. Adapt for secondary platforms
  3. Stagger timing only when the user needs help with ordering

Unless useful, don't add cross-platform references. Most of the time, posts should stand on their own.

Banned Patterns

Delete and rewrite any of the following:

  • "Excited to share"
  • "Here's what I learned"
  • "What do you think?"
  • "link in bio" unless it actually is
  • Generic "professional insights" paragraphs that aren't in the original material

Output Format

Return:

  • Primary platform version
  • Adapted variants for each requested platform
  • Brief explanation of what changed and why
  • Any publishing constraints the user still needs to resolve

Quality Gate

Before delivery:

  • Each version reads like the same author wrote it under different constraints
  • No platform version feels padded or neutered
  • No verbatim copy-pasted copy across platforms
  • Any extra background added for LinkedIn or newsletters is genuinely necessary

Related Skills

  • brand-voice for reusable voice capture derived from material
  • content-engine for voice capture and material shaping
  • x-api for X publishing workflows

Related Skills