tuwen

Marks the deprecated Era multi-page graphic-text workflow and redirects requests to the fengshui video skill.

Updated Jul 8, 2026
One-click install
npx skills add https://github.com/bibidu/era --skill tuwen-bibidu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tuwen
Source: https://github.com/bibidu/era/tree/main/.agents/skills/tuwen
Command: npx skills add https://github.com/bibidu/era --skill tuwen-bibidu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? This Skill prevents agents from accidentally running the retired Era multi-page graphic-text and cover-image production pipeline, and tells them what to do instead when users mention the old workflow. ## Core Features & Use Cases - Deprecation Guard: Blocks any attempt to generate multi-page graphic-text pages, Swiss-style covers, or stacked-text images through the old tuwen pipeline. - Intent Routing: Redirects fengshui, yangzhai, voiceover, or Douyin-link video requests to the fengshui vertical-video skill. - Use Case: A user says "use the tuwen skill to make a fengshui cover set". The agent responds that the graphic-text pipeline is deprecated and either routes the request to the fengshui video skill or declines to produce multi-page graphics. ## Quick Start Tell the agent that the tuwen graphic-text skill is deprecated and ask it to route any fengshui or voiceover video request to the fengshui skill instead.

Frequently Asked Questions about tuwen

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

FAQPage Schema
Is the tuwen graphic-text skill still usable?

No, the tuwen skill is fully deprecated and must not be used to generate multi-page graphic-text pages or covers. The frontend graphic editor at ?tab=graphic still opens for preview, but the agent production pipeline is shut down.

What should I use instead of tuwen for fengshui content?

Use the fengshui vertical-video skill for fengshui, yangzhai, voiceover, or Douyin-link video requests. It produces a 1080x1920 vertical video with cloned voiceover and subtitles instead of static graphic pages.

Can I still run generate-cover.mjs for cover images?

No, generate-cover.mjs and the gc-minimal stacked-text approach are historical leftovers in the repository. They must not be treated as the current production path for covers or graphic-text output.

What happens if a user asks for multi-page graphic output?

The agent should decline and explain that the graphic-text pipeline is deprecated. Only fengshui-related requests are redirected, to the fengshui vertical-video skill; all other multi-page graphic requests are not fulfilled.