kaltstart-triage
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of markdown instructions for legal triage. No malicious code, obfuscation, hardcoded credentials, or persistence mechanisms were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data provided by the user, creating a standard attack surface for indirect prompt injection.
- Ingestion points: The instructions guide the agent to read "Material," "Verträge" (contracts), and "technische Nachweise" (technical proofs) provided during the interaction (SKILL.md).
- Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore potentially malicious instructions embedded within the analyzed documents.
- Capability inventory: The skill implies capabilities for reading files and generating text-based output (e.g., "ergebnis.md").
- Sanitization: The instructions do not specify any validation or sanitization steps for the external content before processing.
Audit Metadata