dsv-aufnahme-statusinformation
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of markdown instructions and does not include any executable scripts, binaries, or environment configuration files.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its primary function is to ingest and process untrusted external reports concerning data breaches.
- Ingestion points: The skill processes user-supplied incident documentation and reports ("Material vorliegt", "gemeldeter Datenschutzvorfall").
- Boundary markers: There are no explicit delimiters or specific "ignore embedded instructions" warnings defined to separate user-provided content from the agent's core instructions.
- Capability inventory: The skill's functionality is limited to generating textual outputs (status reports, matrices, and memos) and does not involve subprocess execution, file system modifications, or network-based exfiltration tools.
- Sanitization: No input validation, escaping, or filtering mechanisms are specified for the material being processed.
Audit Metadata