privacy-notice-generator

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The notice_compliance_checker.py script ingests untrusted text from user-specified files to perform compliance audits. This creates a surface for indirect prompt injection if an attacker embeds malicious instructions within a privacy notice file.
  • Ingestion points: scripts/notice_compliance_checker.py reads content from the file path provided in the notice_file argument.
  • Boundary markers: No delimiters or isolation instructions are present to separate user-provided content from the agent's logic.
  • Capability inventory: The script is restricted to reading local files and printing regex match results; it lacks network, file-write, or subprocess execution capabilities.
  • Sanitization: Input text is processed using standard regex without sanitization, though the impact is limited by the script's lack of dangerous capabilities.
  • [DATA_EXPOSURE]: The audit tool allows the agent to read local file content to verify keywords. While this functionality is intended for privacy notice review, it could be used to read other filesystem data if the agent is misdirected. However, the script does not include any data exfiltration logic.
  • [EXTERNAL_DOWNLOADS]: The skill includes references to official European and UK government domains (e.g., bfdi.bund.de, cnil.fr, ico.org.uk) for legal reference. These are trusted domains and do not involve the download of executable code or binary assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 10:33 AM