eu-ai-act-transparency-assessor-oliver-schmidt-prietz

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill contains instructions for the agent to 'run quietly' when performing web searches and to ask questions 'one at a time'. These are identified as functional UI/UX instructions designed to manage the agent's output volume and interaction flow rather than attempts to bypass safety filters or conceal malicious intent.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes an intake mechanism that ingests an 'ASSESSMENT CONTEXT' block from external sources or other skills. This creates an attack surface for indirect prompt injection; however, the risk is mitigated by explicit instructions to 'echo back' relying facts to the user for validation and to 'flag inconsistencies' before performing any analysis.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes search tools to verify the status of official EU regulatory documents, such as the Digital Omnibus and Commission Guidelines. These activities are limited to retrieving public legal information from well-known authoritative domains (e.g., eur-lex.europa.eu, consilium.europa.eu) and are essential for providing current regulatory guidance.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:32 PM
Security Audit — agent-trust-hub — eu-ai-act-transparency-assessor-oliver-schmidt-prietz