usage-ia-agents-etat

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary purpose is to educate users on the safe and legal use of AI in a professional administrative context. It does not contain any executable code, scripts, or network exfiltration patterns. All instructions are aimed at protecting sensitive data and ensuring human oversight.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user descriptions of their tasks to provide advice on tool selection and data handling. While this creates a natural interface for interaction, the skill explicitly warns about AI hallucinations and mandates human validation of all outputs. (Analyzed as safe due to the absence of dangerous capabilities).
  • [EXTERNAL_DOWNLOADS]: The documentation references official French government resources (e.g., ia.numerique.gouv.fr, assistant.numerique.gouv.fr) and a vendor-specific installation method. These resources are from trusted domains and do not involve unauthorized code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 02:48 PM
Security Audit — agent-trust-hub — usage-ia-agents-etat