avatar-spokesperson-production

What it does

Use this skill to plan, direct, review, and package synthetic presenter videos where a humanlike avatar or digital spokesperson speaks to the viewer. It is provider-independent: adapt the workflow to the avatar, TTS, dubbing, lip-sync, compositing, or platform tools available in the current environment.

Do not treat avatar production as only a prompt-writing task. The core production risks are identity rights, audience transparency, claims accuracy, performance credibility, brand trust, localization, and QA.

How it works

Before generation, confirm the five mentioned non-negotiable production boundaries and the rights/consent review sections:

• Right/licensing checks for avatar and voice across uses, audience, territory, duration, channels, languages, and editing scope.
• Ensure the video will not imply endorsement by a real person or entity unless true and approved.
• Plan required disclosures and platform labels.
• Validate objective/high-risk claims with substantiation and approvals.
• Prepare script, pronunciation list, performance direction, wardrobe/background, captions, and delivery variants.

The first-pass requires filling a YAML-style project brief and a persona brief with fields like goal, audience, format, channels, duration_targets, risk_level, and approval workflow. It also specifies a structured rights/consent review, including minimum consent packets and explicit prohibitions.

It includes a comprehensive script and persona brief section:

• Provide a detailed persona_brief with role, audience relationship, credibility basis, warmth, formality, pace, energy, eye_line, gestures, wardrobe, background, and a must-not-read line.
• Script rules for timing, sentence clarity, contractions, pronunciation marking, and explicit guidance on pacing and pronunciation markers.

Casting and performance direction are defined to ensure comprehension and trust, with criteria on apparent credentials, alignment with identity, and avoidance of deception. Performance direction sections specify eye-line, gesture density, facial expression, head movement, pacing, shot type, and interaction with graphics, plus guidance to break long scripts into takes and to produce localization proofs.

Wardrobe and background specify brand-fit constraints and visual considerations to minimize rendering artifacts and avoid implying unauthorized affiliations. It also covers brand application to backgrounds and captions, ensuring current, approved displays of prices and legal footers.

The skill includes a detailed claims, endorsements, and sensitive-content review with a sample YAML inventory, and escalation rules for high-risk or unverified statements.

Localization and pronunciation guidance emphasize using approved translations, creating pronunciation tables for brand terms, and checking locale-specific formality, units, currency, and gestures.

This skill emphasizes that disclosures are part of the creative plan, not an afterthought, and provides platform-specific labeling recommendations and volatile policy notes for major platforms.

When to use it

Use when planning synthetic presenter videos that require rights management, disclosures, localization, QA, and packaging for delivery. Triggered contexts include corporate explainers, training modules, sales/support avatars, localization spokespeople, product announcements, compliance videos, or social avatar clips.

What it can touch

Declared tools: claude-code, codex, copilot, cursor. The workflow references script creation, localization work, claims inventories, and rendering directions, all of which would involve these tools if executing the steps inside a compliant environment.

Caveats

This skill outlines non-legal guidance and requires escalation for regulated content, minors, political content, or any scenario involving real-person likenesses without explicit approvals. It emphasizes consent packets, explicit prohibited uses, and security requirements for source recordings and generated media. It also notes platform policy volatility for disclosures on YouTube, TikTok, and Meta as of 2026-07-11.