analise-semiotica
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data in the form of uploaded images, videos, and text justifications, which are then used to influence the agent's analysis and the subsequent use of a search tool.
- Ingestion points: User-provided media and justification text are processed in the 'tarefa' section of SKILL.md.
- Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between user-supplied data and its own analysis logic, nor are there instructions to ignore instructions embedded within the media or text.
- Capability inventory: The skill utilizes the
google_searchtool to provide external links for study recommendations based on its analysis. - Sanitization: There is no mention of input sanitization, validation, or filtering for the data ingested from the user.
Audit Metadata