dsv-meldung-ldi-nrw
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill references public contact information for the LDI NRW (e.g., poststelle@ldi.nrw.de) and official government domains (ldi.nrw.de, gesetze-im-internet.de). No sensitive credentials, private keys, or unauthorized network operations were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external documents such as contracts, administrative acts, and expert reports. While this establishes a potential surface for indirect prompt injection if those documents contain malicious instructions, the skill does not include automated tool execution or capabilities that would allow for high-impact exploitation.
- Ingestion points: External documents like 'Verwaltungsakte', 'Vertragsurkunden', and 'Schriftsätze'.
- Boundary markers: None identified in the provided text.
- Capability inventory: Text drafting and legal analysis (no automated file system writes or network exfiltration tools defined within the skill content).
- Sanitization: None specified for external content ingestion.
Audit Metadata