beweisforensik-versionen
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured legal workflows and references well-known official legal portals (e.g., gesetze-im-internet.de, eur-lex.europa.eu) for norm verification.
- [NO_CODE]: No executable scripts, binaries, or command-line tools are included in the skill, which eliminates risks associated with remote code execution or privilege escalation.
- [INDIRECT_PROMPT_INJECTION]: The skill involves analyzing external data such as prompt logs and metadata. 1. Ingestion points: Documents and evidence listed in the 'Beweise anfordern' section. 2. Boundary markers: None specified. 3. Capability inventory: Limited to analytical text generation. 4. Sanitization: None present. This represents a standard surface for an analysis-focused agent.
Audit Metadata