DELLIGHT Content & Marketing Operations

PassAudited by ClawScan on May 1, 2026.

Overview

This is a coherent instruction-only marketing skill, but users should keep human review over public LinkedIn activity, voice-clone use, and third-party AI provider uploads.

This appears safe as a planning and drafting skill. Before installing, decide whether the agent may only draft content or may also publish/reply on LinkedIn. Keep approval controls for public posts, confirm permission for any cloned voice use, and avoid uploading confidential materials to third-party AI services unless your policy allows it.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If used with access to a social account, agent-generated replies could affect the company or founder’s public reputation.

Why it was flagged

This is guidance for public social-media engagement. It is aligned with marketing operations, but the artifacts do not state an approval step before posting replies.

Skill content
- Reply to every comment within 2 hours
Recommendation

Use the skill to draft and plan social content, and require human approval before publishing posts or replies.

What this means

Generated media could appear to represent a real person’s voice or presence, which may create reputational or consent concerns.

Why it was flagged

The skill contemplates synthetic voice use for a named person. This is purpose-aligned for branded marketing videos, but voice cloning can affect audience trust if not authorized or appropriately disclosed.

Skill content
**Conversational voice** — Arthur's cloned voice or warm American male
Recommendation

Use cloned voice assets only with clear authorization and follow any disclosure, platform, and consent requirements.

What this means

Company, customer, or campaign information could be processed under third-party provider policies if included in prompts or uploaded assets.

Why it was flagged

The skill names external AI and media-generation providers. This is expected for video and marketing production, but prompts, assets, or brand materials may leave the local environment if those services are used.

Skill content
- WaveSpeed API (v3): Image/video generation ... - ElevenLabs: Voice synthesis ... - Gemini: Research ... - GPT Image 1
Recommendation

Avoid sending confidential or customer data to external providers unless approved, and check provider retention and privacy settings.