cinematic-visual-developmentlisted

# Give the model the goal and only the necessary visual anchors Plan what the viewer must see and what must remain consistent. Separate **must preserve**, **may change**, and **model discretion** in the shot contract. Text direction is a valid starting point for clothing, locations, action, lighting and camera. An image is needed when it provides a specific control, not merely because a subject or location appears in the script. Read `minimax-h3-video` for route controls and request handling, and `multi-clip-cinematic-generation` for the editorial handoff. Verify the live schema before using an input; endpoint support does not make that input necessary. ## Start with the minimum sufficient references | Subject type | Starting reference when consistency is needed | Add detail when required | | --- | --- | --- | | Human / presenter | One approved face-only identity reference | Exact costume, body details, profile fidelity or a demonstrated identity miss | | Animal / pet | One image showing identifying markings and silhouette | A hidden side or marking that must match | | Mascot / creature | One approved identity/style view | Required silhouette, material or geometry not shown there | | Product / object | A view showing the required geometry and branding | Exact unseen details, scale or multiple required views | | Vehicle / robot | One identifying view | Required shape, markings or action-specific detail | | Environment | Text scene direction | Exact return geography, brand/