higgsfield-mcp

Generate videos and images via the Higgsfield MCP server with cost estimation.

Updated May 24, 2026
One-click install
npx skills add https://github.com/maddoherty23/higgsfield-claude-code --skill higgsfield-mcp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: higgsfield-mcp
Source: https://github.com/maddoherty23/higgsfield-claude-code/tree/main/skills/higgsfield-mcp
Command: npx skills add https://github.com/maddoherty23/higgsfield-claude-code --skill higgsfield-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires higgsfield_mcp_client, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill simplifies the process of creating and retrieving AI-generated video and image assets through the Higgsfield MCP server, offering a streamlined workflow for marketing content creation.

Core Features & Use Cases

  • AI Video Creation: Generate various video formats like UGC, premium product reveals, studio lookbooks, and feature walk-throughs using different Higgsfield models.
  • AI Image Creation: Create still images with options like product showcases, influencer recreations, and custom character creation.
  • MCP Integration: Leverages the Higgsfield MCP server for seamless interaction and credit management without the need for API keys.
  • Cost Estimation: Provides real-time credit cost estimation to prevent overages.
  • Dialogue Confirmation: Ensures dialogue content in video generation is reviewed and approved by the user.
  • Media Management: Handles media uploads and confirms media readiness for generation.
  • Reusable References: Supports the creation and use of reusable characters and reference elements for consistent branding and styling.
  • Multi-Clip Support: Generates multiple video clips when a single clip is too long, ensuring the style's continuity.
  • Stitching: Offers to stitch generated video segments together for a seamless output.

Quick Start

To create a video using the Seedance 2.0 model, start by using the following command: use skill higgsfield-mcp generate_video model=seedance_2_0 prompt="Product showcase video with dynamic transitions and on-screen text" medias=[{value: "product_image.jpg", role: "image"}] duration=15 resolution="720p" aspect_ratio="9:16" generate_audio=true

Frequently Asked Questions about higgsfield-mcp

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

FAQPage Schema
How do I generate AI video content for marketing using different models?

Generate AI video content by specifying a model, prompt, media inputs, and parameters like duration and resolution. The system handles end-to-end video generation, supporting formats like UGC, product reveals, and studio lookbooks.

Can I estimate credit costs before generating AI images and videos?

Yes, you can estimate credit costs before generation. The system provides real-time cost estimation to help manage credits and prevent overages during AI image and video creation.

How do I maintain consistent branding across multiple AI video clips?

Maintain consistent branding by creating reusable characters and reference elements. The system supports multi-clip generation with continuity and can stitch video segments together for a seamless output.

Do I need an API key to create marketing content with the Higgsfield MCP server?

No, you do not need an API key. The system integrates directly with the Higgsfield MCP server for seamless interaction and credit management without requiring API key authentication.

How does media upload work for AI image and video creation workflows?

Media upload works by allowing you to pass local media files as inputs for generation. The system handles the upload process and confirms media readiness before starting the AI creation workflow.

What is the best way to review dialogue content in generated videos?

The best way to review dialogue is through the built-in confirmation step. The system ensures all dialogue content intended for video generation is reviewed and explicitly approved by the user.