performance-direction

What it does

Provides provider-independent guidance to direct synthetic performers across video, avatar/spokesperson clips, animation, ads, film scenes, social content, and voice-led media. It instructs how to plan acting beats, objective/obstacle/action, blocking, gesture, eye-line, facial expression, body language, lip-sync, voice alignment, rehearsal prompts, iteration, continuity, consent, and performance QA.

How it works

Outlines a structured approach for each production task:

• Build a performance brief with medium/constraint, casting, given circumstances, objective, obstacle, action verb, beat turn, visible behavior, vocal behavior, continuity, and safety/rights.
• Direct playable beats by specifying concrete, action-oriented behaviors rather than vague emotions, including explicit changes in what the performer wants, tactics, and observable changes.
• Calibrate emotion and expression by outlining underlying emotion only after the action, then assign a screen intensity from 0 to 5 for faces and observable cues for eyes, brow, mouth, and head.
• Define blocking, gesture, and eye-line with defaults for direct-to-camera, dialogue, product ads, film, and animation; assign a verb to each gesture and ensure eye-line continuity.
• Align voice, lip-sync, and face by detailing intention, pace, pauses, emphasis, breath, and contour; provide provider-specific performance markup when supported; follow a 6-step lip-sync/voice workflow before generation and perform dubbing/localization choices if needed.
• Describe avatar and animation constraints (presenter, talking photo, AI cinematic character, rigged animation, voice-only) and outline direction styles and risks for each medium.
• Address casting, likeness, and consent with a decision tree on performer type and require written scope for custom or recognizable likeness/voice work.
• Rehearse through prompts with a structured rehearsal sequence (table read, pose rehearsal, beat rehearsal, continuity rehearsal, final performance) and provide iteration notes with concrete adjustments.

When to use it

Use when an agent must cast or direct synthetic performers, avatars, animated characters, talking heads, AI video characters, or voice performances; for planning, blocking, and QA, and to ensure consent and provenance considerations are embedded in the workflow.

What it can touch

The skill references tools and scripts for directing avatars and voice performances and implies prompts, beats, and QA steps; it does not specify executable commands or external tooling beyond general direction elements. (No explicit required-tools list beyond high-level guidance provided.)

Caveats

Cites consent, rights, and provenance concerns for synthetic performance, including likeness and voice-use terms; emphasizes that capabilities and timing may vary by provider and project, and that constant attention to continuity and QA is required. No guarantees about model output quality or copyright outcomes are stated.