810-regulations-eu-mifid-ii

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill does not contain any detected malicious patterns, obfuscation, or unauthorized commands.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The instructions emphasize the importance of identifying and redacting sensitive information like API keys, credentials, and private keys, replacing them with a specific placeholder ([REDACTED_SECRET]). No network exfiltration patterns were found.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external project data, which is a potential vector for indirect prompt injection. However, it mitigates this risk by instructing the agent to treat such data as evidence rather than instructions and requiring explicit evidence-backed answers for its questionnaire.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 02:07 PM
Security Audit — agent-trust-hub — 810-regulations-eu-mifid-ii