Influencer

ReviewAudited by ClawScan on May 10, 2026.

Overview

This instruction-only skill is mostly coherent for creating AI influencers, but it includes AI-detection-evasion guidance and public monetization/posting workflows that need careful review.

Use this skill only if you intend to manage clearly disclosed fictional AI personas. Ignore or remove the 'Avoiding Detection' guidance, verify consent/licensing for any likeness or voice source, keep API keys and the ~/influencers folder private, and require your approval before any public posting, brand communication, contracts, or payment-related actions.

Findings (4)

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.

Medium

#ASI09: Human-Agent Trust Exploitation

What this means

The agent could help create synthetic influencer content that is harder for viewers or platforms to recognize as AI-generated.

Why it was flagged

The skill explicitly teaches reducing detectability of AI-generated images, which can undermine audience/platform transparency even though another file recommends disclosure.

Skill content

## Avoiding Detection

AI-generated images can be detected. Reduce risk:

1. **Add subtle imperfections**

Recommendation

Treat disclosure as mandatory, remove or ignore detection-evasion guidance, and require visible AI labels on generated influencer content.

Low

#ASI02: Tool Misuse and Exploitation

What this means

If connected to social, email, or payment tools, the agent could affect public accounts or business relationships.

Why it was flagged

The workflow reaches external business communications, contracts, payments, and public posting; this is aligned with monetization but should not be automated without user approval.

Skill content

Brand Deal Workflow

1. **Receive inquiry** → Respond within 24h ... 3. **Negotiate terms** ... 4. **Create contract** ... 8. **Post on agreed date**

Recommendation

Require explicit user confirmation before outreach, negotiation, contract changes, payment handling, scheduling, or posting.

Low

#ASI03: Identity and Privilege Abuse

What this means

Leaking or mishandling the API key could allow unauthorized use of the user's ElevenLabs account.

Why it was flagged

The voice-generation example uses an ElevenLabs API key; this is expected for TTS/voice cloning but is still a credential that grants provider-account access.

Skill content

-H "xi-api-key: $ELEVENLABS_API_KEY"

Recommendation

Store API keys securely, do not paste them into shared chats or files, and revoke/rotate keys if exposed.

Low

#ASI06: Memory and Context Poisoning

What this means

Private campaign data, generated identities, analytics, and tool identifiers may persist across sessions and be reused in later outputs.

Why it was flagged

The skill relies on persistent local persona, reference, scheduling, analytics, and tool-configuration files that can influence later agent work.

Skill content

Each influencer lives in a dedicated folder: `~/influencers/` ... `identity.md` ... `reference/` ... `schedule.md` ... `analytics.md` ... `tools.md`

Recommendation

Keep the workspace private, review files before reuse, and avoid storing secrets or real-person likeness/voice data without consent.