ads-photoshoot
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No hardcoded credentials, malicious obfuscation, or unauthorized network activity was detected.
- [SAFE]: No persistence mechanisms, privilege escalation, or dynamic execution patterns are present.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection through external data ingestion. 1. Ingestion points: 'authorized source image' and 'validated brand profile' as defined in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Image generation and file-write operations to store assets and manifests. 4. Sanitization: Mandatory human review and validation of product fidelity and policy compliance.
Audit Metadata