811-regulations-eu-market-abuse-regulation

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a documentation-only toolkit consisting of markdown instructions, questionnaires, and report templates. It does not contain any executable scripts, binary files, or automated tool invocations.
  • [DATA_EXPOSURE]: The skill demonstrates strong security practices by explicitly instructing the agent and user to redact all secrets, credentials, and sensitive private information (using markers like [REDACTED_SECRET]) before generating any reports or summaries.
  • [EXTERNAL_DOWNLOADS]: While the skill references official EU regulatory links (EUR-Lex), it contains explicit instructions ('SOURCE-FIRST REVIEW') forbidding the agent from fetching or ingesting external web pages at runtime, ensuring it only uses the provided local reference material.
  • [PROMPT_INJECTION]: No patterns associated with prompt injection, jailbreaking, or instruction override were found. The instructions are consistently focused on compliance auditing and engineering controls.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 05:43 PM
Security Audit — agent-trust-hub — 811-regulations-eu-market-abuse-regulation