goose-product-photos
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions include positive security practices, such as explicitly directing the agent never to ask for keys, tokens, passwords, or account identifiers.
- [SAFE]: The skill implements a financial guardrail by requiring the agent to perform a dry run for pricing and obtain explicit user confirmation before proceeding with paid photo generation.
- [SAFE]: The execution flow is strictly defined through a set of specialized tools (e.g., brand_knowledge_browse, brand_read, photos_generate), which limits the agent's actions to the intended service scope.
- [SAFE]: The persona-based constraints (e.g., not mentioning technical IDs or JSON) are designed for user experience and do not attempt to bypass safety filters or hide malicious activity.
- [SAFE]: No signs of obfuscation, remote code execution, or unauthorized data exfiltration were detected in the instructions or metadata.
Audit Metadata