Seedance 2.5 Tutorial: Full AI Commercial Workflow (Claude + MCP)
AI filmmakers often struggle with visual continuity and time-consuming pre-production; this workflow shows how connecting Claude to Higgsfield MCP and Milanote automates art direction and metadata extraction to maintain stylistic consistency.
What this lesson covers
Rourke details how he produced a 4K holiday heist commercial in five hours using Seedance 2.5 and an automated pre-production workflow. He shows how to link Claude with Higgsfield MCP and Milanote to automatically construct shot breakdowns, character sheets, and cinematography references.
Key takeaways from the creator
AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.
- 01:48 ↗
Claude can be configured to structure pre-production boards in Milanote, populating character sheets, music cue timestamps, and scene-by-scene breakdowns.
- 03:32 ↗
Pulling real-world cinematography metadata such as exact color hex codes, camera bodies, and lens types from ShotDeck establishes strict continuity across generated scenes.
- 05:01 ↗
Integrating Claude with Higgsfield MCP allows the agent to generate custom location assets that match the color grading and aesthetic of ShotDeck reference imagery.
- 06:42 ↗
Higgsfield MCP can be paired directly with Claude through Higgsfield's connector dashboard or inside Claude's customization settings.
Workflow outlined in the video
- Set up a Milanote board template with dedicated sections for character sheets, shot descriptions, dialogue, and cinematography references.
- Connect Higgsfield MCP to Claude using either the direct web authorization or Claude's desktop integration panel.
- Reference visual databases like ShotDeck to identify specific camera models, focal lengths, and hex palettes for the target visual style.
- Provide Claude with the Milanote board link alongside a narrative concept prompt to auto-populate the shot list and production notes.
- Trigger image generation through Claude via Higgsfield MCP to produce cohesive set and location reference assets.
Before you use this workflow
These notes describe the source video at its publication date. Model access, pricing, connectors and interfaces may have changed. Check the original source and the provider’s current documentation before installing an add-on, connecting an account or spending credits.
ReelStack has not independently tested this workflow. Preview one representative shot and check motion, continuity and output quality before applying it to a full production. No result, cost saving or model capability is guaranteed.
Explore the tools
How to interpret the numbers
31,585 views captured 21 November 2026. ReelStack recommendation score: 47/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.
Show the calculation
- 65% performance: 0.64× views vs 8 other channel videos observed at a similar age. View-evidence factor 97% (views / (views + 1,000))
- 25% freshness: 71/100 with a 14-day half-life
- Momentum pending: collecting daily snapshots; its 20% weight goes to observed performance, not free points
- 10% engagement: 44/100; likes + 4× comments, smoothed with a 500-view neutral prior