forschungszulage
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from the user regarding their research projects to facilitate the drafting of formal application texts. \n
- Ingestion points: The skill actively prompts users to provide detailed information about their research objectives, methodologies, and technical risks as specified in the 'Neuen Antrag erstellen' workflow in
SKILL.md. \n - Boundary markers: There are no explicit delimiters or specific instructions for the AI to ignore potential commands embedded within the user's project descriptions. \n
- Capability inventory: The skill does not have access to tools for command execution, file writing, or network operations, which limits the potential for harm from malicious user input. \n
- Sanitization: The instructions do not define any sanitization or validation steps for the user-provided project data before it is processed by the agent.
Audit Metadata