fremdmaterial-check
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process "IP Content", "Arbeitsergebnisse" (work results), and various legal/financial documents like contracts, invoices, and status applications.
- Ingestion points: The workflow processes user-provided text and external documents provided by the client as listed in the "Dokumente und Beweismittel" section.
- Boundary markers: The instructions do not define specific delimiters or safety warnings to ignore instructions that might be embedded within the processed documents.
- Capability inventory: While the skill itself does not define dangerous commands, it is designed to use a "Live-Check" capability to browse official websites, creating a potential path for processing malicious instructions from external web content.
- Sanitization: There are no explicit instructions to sanitize or escape the content of the documents or web data before analysis.
Audit Metadata