brand-dev
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external websites.\n
- Ingestion points: External website content is analyzed and returned to the agent context via the
/v1/brand/ai/query,/v1/brand/ai/products, and/v1/brand/styleguideendpoints (defined in SKILL.md).\n - Boundary markers: The skill does not specify the use of delimiters or instructions to the agent to disregard potential instructions embedded in the crawled content.\n
- Capability inventory: The skill uses the
orthtool to interact with the Brand.dev API, which includes features like screenshot capture and AI-driven data extraction.\n - Sanitization: There is no mention of sanitization or filtering of the retrieved web content before it is processed by the AI.
Audit Metadata