content-pipeline

Generate a multi-stage content pipeline with review, rerun, and rollback per node.

10|2|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/PPPPanda/contentPipe --skill content-pipeline-ppppanda
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: content-pipeline
Source: https://github.com/PPPPanda/contentPipe/tree/main
Command: npx skills add https://github.com/PPPPanda/contentPipe --skill content-pipeline-ppppanda

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ContentPipe helps you turn a vague topic into a complete, multi-stage AI-generated article pipeline (scout → research → write → image plan → image gen → format → publish), with per-node review, rerun, and rollback.

Core Features & Use Cases

  • Interactive multi-node pipeline: pause for review, continue discussion, and rerun specific stages without restarting everything.
  • Evidence-first research: generates structured research packets for safe writer consumption (fact-check and sourcing discipline).
  • Director-driven visual planning: produces a visual plan that maps image placements to article sections for precise insertion.
  • WeChat-compatible formatting: converts Markdown into WeChat-safe inline-styled HTML and supports dark-mode safe template styling.
  • Publishing/export orchestration: drafts and exports content for target platforms (e.g., WeChat draft saving; XHS outputs as JSON package).

Quick Start

Tell the AI: 帮我写一篇关于 AI Agent 的公众号文章,按“可审核的七节点流水线”生成并给出每一步的审核产物。

Frequently Asked Questions about content-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a WeChat article from a topic with a reviewable content pipeline?

A reviewable content pipeline for WeChat articles turns a vague topic into a multi-stage workflow producing structured artifacts for scouting, research, writing, and formatting. You can pause for review, rerun specific stages, and rollback without restarting everything.

What is evidence-based research in an AI article writing workflow?

Evidence-based research in an AI article writing workflow generates structured research packets with fact-check and sourcing discipline for safe writer consumption. This ensures the content pipeline produces verifiable, machine-readable outputs before the writing stage begins.

How do I convert Markdown to WeChat-safe inline-styled HTML?

To convert Markdown to WeChat-safe inline-styled HTML, use a formatting node within the content pipeline that supports dark-mode safe template styling. This step ensures the final article output is compatible with WeChat's publishing requirements.

Can I rerun specific stages of an AI content pipeline without restarting the whole workflow?

Yes, you can rerun specific stages of an AI content pipeline without restarting the whole workflow. The pipeline supports per-node session continuity via OpenClaw Gateway, allowing you to pause for review, continue discussion, and rollback to previous stages as needed.

Does the multi-stage content pipeline work for exporting content to platforms other than WeChat?

Yes, the multi-stage content pipeline supports exporting content to platforms other than WeChat. It handles publishing and export orchestration for target platforms, generating outputs like JSON packages for XHS and saving drafts for WeChat.

What are the limitations of using a deterministic file-commit contract for content generation?

Using a deterministic file-commit contract for content generation requires strict validator-based artifact generation to ensure YAML, JSON, and Markdown outputs remain machine-readable. This constraint means the pipeline relies on structured formats rather than free-form text generation.