urlaubsanspruch
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied employment history and health data to generate reports and legal drafts, creating a potential surface for indirect prompt injection.\n- Ingestion points: Input data including employment dates, vacation balances, and sick leave periods enter the agent context via the "Eingaben" section of SKILL.md.\n- Boundary markers: The skill does not employ specific delimiters or specialized system instructions to isolate user-provided data from the processing logic.\n- Capability inventory: The skill orchestrates sub-agents (researcher, drafter, reviewer) to generate legal content; however, no direct system-level tool access or network capabilities are requested in the skill's frontmatter.\n- Sanitization: There is no evidence of validation or sanitization routines for the data provided by the user within the skill's instructions.
Audit Metadata