co-op-game-intro-generator

Generates two-player co-op game intro videos via confirmation image and Minimax H3 prompts.

38|4|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/Qo-qiao/ComfyUI-omni-llm --skill co-op-game-intro-generator-qo-qiao
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: co-op-game-intro-generator
Source: https://github.com/Qo-qiao/ComfyUI-omni-llm/tree/main/skills/co-op-game-intro-generator
Command: npx skills add https://github.com/Qo-qiao/ComfyUI-omni-llm --skill co-op-game-intro-generator-qo-qiao

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Creating a polished two-player co-op game opening video requires coordinating character identity, menu UI design, color systems, and video event timing, which is difficult to get right in a single generation pass. ## Core Features & Use Cases - Style-First Confirmation Workflow: Collects player names, game title, and a chosen visual style, then generates one confirmation image from a fixed menu framework before any video is produced. - Identity-Locked Characters: Uses optional player reference images only for identity anchors (face silhouette, hairstyle, proportions) while re-rendering faces in the selected visual style. - Template-Driven H3 Video Generation: Refills a fixed timeline video prompt template (menu, equipment configuration, world loading, gameplay entry) and generates the final video with Minimax H3. - Use Case: Two friends want a stylized intro video for their co-op game concept with their names on player cards and a console-style main menu; the Skill confirms the menu image first, then produces the full opening animation. ## Quick Start Ask the assistant to create a co-op game intro video, then choose a visual style and provide two player names, a game title, and optional character reference images.

Frequently Asked Questions about co-op-game-intro-generator

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

FAQPage Schema
How do I create a co-op game intro video with player names?

Provide two player names, a game title, and a visual style. The Skill first generates a single confirmation image of the game main menu, and after your approval it refills the video prompt template and generates the final intro video with Minimax H3.

How does the confirmation image workflow work before video generation?

The Skill builds one confirmation image from a fixed 16:9 menu framework with palette-linked UI, buttons, icons, and typography. No video is generated until you approve the image; any style or name changes send it back to the image prompt step.

Can I use my own photos as character references?

Yes, you can upload reference images for PLAYER 1 and PLAYER 2. They are used only for identity anchors like face silhouette, hairstyle, and proportions; the faces are re-rendered in your chosen visual style without photographic realism.

Does this Skill work outside the MiniMax Hub agent?

No. The Skill requires the MiniMax Hub agent with canvas workspace and Minimax H3 video generation, and it is not portable to generic agent harnesses.

What are the limitations of this game intro generator?

It is not suited for playable game development, complex multi-page UI, exact brand-logo replication, or generic character-free title sequences. It focuses on a single two-character menu intro video.

Why do characters swap identities or faces drift in the generated video?

Identity swaps happen when name, position, and color cues are weak. The fix is to strengthen left/right positioning and per-player colors, reuse the uploaded reference images, and explicitly preserve identity, hairstyle, and outfit anchors.