skills/klotzkette/claude-fuer-deutsches-recht/adressatenkreis-bundestag-bundesregierung/Gen Agent Trust Hub
adressatenkreis-bundestag-bundesregierung
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external documents such as activity reports and financial disclosures, which creates a surface for indirect prompt injection.
- Ingestion points: Documents and evidence gathered during the 'Arbeitsweg' phase (SKILL.md), including 'Lobbyregistereintrag' and 'Tätigkeitsbericht'.
- Boundary markers: The instructions lack delimiters or isolation markers to distinguish external data from the skill's instructions.
- Capability inventory: The skill defines a logical reasoning workflow but does not include any scripts or automated tool-invocation instructions for file-system or network access.
- Sanitization: There are no instructions for sanitizing or filtering the content of external documents before processing.
Audit Metadata