flowy-brandkit
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data fetched from external websites provided by users via the create_brandkit tool. This data (e.g., brand voice, keywords) is intended to be incorporated into subsequent generation prompts, which can be a vector for indirect instructions.
- Ingestion points: Data is pulled from external URLs provided to the create_brandkit tool in SKILL.md.
- Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from obeying instructions embedded in the extracted brand data.
- Capability inventory: The skill uses tools to read, create, and update brand kits, and specifically advises weaving extracted content into future generation prompts.
- Sanitization: There is no mention of sanitization or filtering for the content extracted from external websites before it is used in prompts.
Audit Metadata