data-act-anwendungsbereich
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from users, such as product descriptions and contract details, to perform legal assessments.
- Ingestion points: Input fields for 'Produkt- oder Dienstbeschreibung' and 'Zeitpunkt des Inverkehrbringens' in SKILL.md.
- Boundary markers: The skill does not specify the use of delimiters or 'ignore embedded instructions' warnings for external input.
- Capability inventory: The skill architecture involves sub-agents drafting responses and calling a local Python script (scripts.legal_calc.cli) to calculate deadlines.
- Sanitization: No explicit input validation or escaping mechanisms are documented for the facts processed by the agents.
- [DYNAMIC_EXECUTION]: The skill utilizes a local Python module to perform deterministic legal date arithmetic.
- Evidence: The 'Deterministische Berechnung' section of SKILL.md instructs the agent to execute 'python -m scripts.legal_calc.cli' with arguments derived from the fact pattern.
Audit Metadata