mandatsannahme-gwg
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from various communication channels.
- Ingestion points: Processes data from E-mail, beA (special electronic lawyer mailbox), physical letters, messenger services, and file uploads as specified in Step 1 of the Guided Process.
- Boundary markers: The skill body lacks instructions for the agent to use delimiters or guardrails when processing these external inputs to prevent the agent from following instructions embedded within the data.
- Capability inventory: The skill generates multiple document templates and initiates data transfers to other internal modules such as the deadline monitor and accounting tools.
- Sanitization: There are no documented procedures for sanitizing or validating the content of external inputs before they are used to generate legal output documents.
- [EXTERNAL_DOWNLOADS]: The skill provides references and links to official German legal and regulatory sources for money laundering prevention.
- The skill directs users to the Bundesrechtsanwaltskammer (BRAK) website for money laundering prevention guidelines.
- It references the official German federal law portal (Gesetze im Internet) for the latest Money Laundering Act (GwG) text.
- It points to the official portal of the Financial Intelligence Unit (FIU) for money laundering reporting guidelines.
Audit Metadata