crisis-detector

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted user data describing crisis scenarios, which represents a potential surface for indirect prompt injection.
  • Ingestion points: User-provided text descriptions in crisis assessment prompts (SKILL.md).
  • Boundary markers: None identified within the provided examples or instructions.
  • Capability inventory: The skill contains no executable scripts, shell commands, or network-capable tools.
  • Sanitization: No sanitization of user-provided crisis data is performed.
  • [SAFE]: No malicious behavior, obfuscation, or dangerous command execution patterns were found. The skill consists entirely of instructional markdown content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:26 PM
Security Audit — agent-trust-hub — crisis-detector