turbo-zeitnarrative
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 analyze documents from external sources, including opposing parties, which could contain hidden instructions to bias or manipulate the agent's output.
- Ingestion points: Processes external "Verwaltungsakte" (administrative files), "Vertragsurkunden" (contracts), and "Schriftsätze" (pleadings) as defined in the 'Arbeitsweg' and 'Triage' sections of SKILL.md.
- Boundary markers: The instructions lack specific requirements for delimiters or 'ignore' tags when processing external text.
- Capability inventory: The skill generates formal legal drafts and interacts with research modules; no direct dangerous capabilities (like shell execution) are defined in this file.
- Sanitization: Mentions a 'Datenschutzcheck' for PII redaction, but does not provide logic for sanitizing instruction leakage from untrusted documents.
Audit Metadata