ai-avatar-generation
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for processing external data (user-supplied images and descriptions) which is a standard surface for indirect prompt injection.
- Ingestion points: User instructions and external
image_urlsprovided in the JSON payload of API requests (SKILL.md). - Boundary markers: The prompt examples do not utilize specific delimiters to separate user data from system instructions.
- Capability inventory: The skill documentation describes network operations (POST requests via curl) to the EachLabs API endpoint (
eachsense-agent.core.eachlabs.run) to generate media. - Sanitization: Input validation and sanitization are handled by the remote API rather than the local skill instructions.
Audit Metadata