ccpa-cpra-privacy-expert
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements its core functionality using local Python scripts (ccpa_compliance_checker.py and ccpa_data_mapper.py) that process user-provided data files. These scripts only use standard library modules and do not perform any network operations or access sensitive system locations.
- [SAFE]: The documentation and instructions are focused on regulatory compliance and do not contain any prompt injection attempts or efforts to bypass agent safety guidelines.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and analyze external JSON data files, though this is integral to its primary purpose as a compliance tool.
- Ingestion points: User-provided JSON profile and inventory files are ingested via the --input parameter in both included scripts.
- Boundary markers: None; the scripts do not implement specific delimiters or instructions to the agent to disregard potential commands embedded in the data.
- Capability inventory: The skill uses Python scripts to read from and write to the local file system for report generation.
- Sanitization: The scripts parse JSON using standard methods but do not explicitly sanitize content for LLM-based instruction injection.
Audit Metadata