x-post-playbook-skill

Turn X/Twitter content into reusable execution playbooks.

Built by

• Ash 🌿 (OpenClaw) — architecture, content-type engine, engagement analysis, agent-browser integration
• C3 🌙 (OpenClaw) — initial implementation and packaging
• B3 (BlueBirdBack) — owner and maintainer

What this skill does

• Mode A — Fetch any X post and convert it into a structured playbook: thesis, engagement quality, key numbers, why it worked, content-type-aware workflow steps, and one actionable next step
• Mode B — Mine an account's recent posts for dominant patterns (async ops, domain flipping, ship-fast, memory-first, etc.) with per-pattern engagement rates and concrete upgrade suggestions
• Mode C — Explain why a specific post performed well (virality drivers, engagement rate, reply dynamics)

Content types detected automatically

Type	Example signal	Workflow template
commerce	"$4.99", "domain", "sold"	5-step domain/asset flip playbook
announcement	"launched", "shipped", "上线"	test → compare → share
question	ends with ?, "why/how/what"	hypothesis → experiment → publish
workflow	"step", "cron", "agent", "自动"	capture → extract → automate → review
opinion	default	claim → test → verdict

Beginner quick start (English prompts, no code)

Talk to your agent in plain English — no commands needed.

Mode A — One post → playbook

• "What can we learn from this post? <POST_URL>"
• "Turn this into an action playbook: <POST_URL>"
• "Give me the core thesis, workflow steps, automation hooks, and one next step for: <POST_URL>"

Expected output: a structured markdown playbook with engagement quality score, key numbers extracted, and a content-type-aware 5-step workflow.

Mode B — One account → pattern report

• "Learn from @<handle>'s recent posts in the last 7 days."
• "Find recurring patterns from @<handle> and suggest what to add to my workflow."
• "Mine @<handle> for top patterns this week, then give me one action I can do today."

Expected output: dominant patterns ranked by engagement rate, evidence links, and a concrete per-pattern skill upgrade.

Mode C — Why did this post work?

• "Why did this post get <N> views and <M> likes? <POST_URL>"
• "Explain the virality drivers for: <POST_URL>"
• "What made this perform so well? <POST_URL>"

Expected output: engagement rate label (🔥 viral / ⚡ strong / 👍 normal / 📉 low), reply dynamics, format analysis, and a one-paragraph explanation.

More ready-to-use prompts

• "Is this claim verified, mostly plausible, or weak? <POST_URL>"
• "Extract only practical steps. Skip opinions: <POST_URL>"
• "Compare these two posts — where do they agree and disagree? <URL_A> <URL_B>"
• "Turn this into a 7-day execution plan with daily actions: <POST_URL>"
• "Draft a friendly X reply in Chinese for: <POST_URL>"
• "Make a decision checklist from this post: <POST_URL>"

Common mistakes

• Asking too broad ("analyze everything") instead of one clear task
• Not providing a specific URL or handle
• Asking for many outputs at once instead of one focused result

Install in OpenClaw

# Ask your agent:
"Install the x-post-playbook skill from https://github.com/BlueBirdBack/x-post-playbook-skill"

Or clone manually:

git clone https://github.com/BlueBirdBack/x-post-playbook-skill \
  ~/.openclaw/workspace/skills/x-post-playbook
npm install -g agent-browser && agent-browser install --with-deps

Attribution & thanks

• Primary source ideas from public posts by QingYue (@YuLin807):
  • https://x.com/YuLin807/status/2025640139947647480
  • https://x.com/YuLin807/status/2025804235707916626
  • https://x.com/YuLin807/status/2025043042840051931
  • https://x.com/YuLin807/status/2025244702992466402
• Fetch dependencies:
  • agent-browser by Vercel Labs (primary): https://github.com/vercel-labs/agent-browser
  • x-tweet-fetcher by ythx-101 (fallback): https://github.com/ythx-101/x-tweet-fetcher
• Reference analysis artifacts in references/
• Huge thanks to QingYue for sharing workflows openly.