gdpr-dsgvo-expert
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by scanning local codebase files and reading user-provided JSON inputs for compliance reporting. This creates a surface where instructions embedded in these files (e.g., in source code comments or JSON fields) could attempt to influence the AI agent's compliance determination. However, the risk is minimized as the skill lacks dangerous capabilities such as network access, shell execution, or dynamic code evaluation.\n
- Ingestion points:
scripts/gdpr_compliance_checker.py(reads all files in target directories) andscripts/dpia_generator.py(reads processing activity JSON files).\n - Boundary markers: The skill does not implement specific delimiters or instructions to ignore embedded prompts in the ingested data.\n
- Capability inventory: The scripts are limited to file system read/write operations using standard Python libraries; no subprocess, network, or eval/exec functionality is present.\n
- Sanitization: No input sanitization is performed on the ingested content beyond regular expression matching.\n- [SAFE]: The skill's primary functionality is to identify privacy risks and generate documentation. All scripts use standard, built-in Python libraries and do not perform any hidden network operations, remote code execution, or persistence mechanisms. The behavior is consistent with the skill's stated purpose of privacy compliance automation.
Audit Metadata