eu-ai-act-readiness

Pass

Audited by Gen Agent Trust Hub on Aug 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a local Python utility (scripts/check_ai_inventory.py) to automate the structural verification of AI system inventories. This script is built using standard libraries and includes defensive checks to prevent accidental file overwriting.
  • [EXTERNAL_DOWNLOADS]: The skill references multiple official European Union resources, including EUR-Lex and Commission policy sites. These are trusted, well-known domains used for legal and regulatory compliance research.
  • [PROMPT_INJECTION]: The skill processes external CSV files containing user-supplied AI system data. This constitutes an indirect prompt injection surface as the agent is expected to review the content of these files.
  • Ingestion points: User-provided inventory CSV files referenced in the workflow.
  • Boundary markers: The skill documentation specifies that tool outputs are for structural screening only, but lacks explicit delimiters for the processed CSV data itself.
  • Capability inventory: Local command execution (python3) and file system read/write access.
  • Sanitization: The validation script checks for schema compliance and keywords but does not sanitize or escape narrative content intended for agent review.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 9, 2026, 08:37 PM
Security Audit — agent-trust-hub — eu-ai-act-readiness