skills/klotzkette/claude-fuer-deutsches-recht/it-prozess-versandmappe-endfertigen/Gen Agent Trust Hub
it-prozess-versandmappe-endfertigen
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill involves the ingestion and processing of external data sources (tickets, logs, and emails) which may contain adversarial instructions targeting the agent.
- Ingestion points: Sections 1 and 3 define a process for reading and converting external tickets, technical logs, and emails.
- Boundary markers: The instructions do not include delimiters or warnings to the agent to disregard instructions within these external files.
- Capability inventory: The skill lacks high-risk capabilities such as network exfiltration or shell execution tools.
- Sanitization: The skill contains no explicit requirements for sanitizing or escaping external content before it is processed by the agent.
Audit Metadata