csv-processor

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill defines patterns for ingesting and processing external CSV data, which introduces a surface for indirect prompt injection.
  • Ingestion points: Untrusted data enters the context through the import_contractors FastAPI endpoint and the parseContractorCsv TypeScript function via file uploads.
  • Boundary markers: While the skill enforces structural integrity using Zod and Pydantic, it does not implement specific prompt boundaries or instructions to treat the data as non-executable text.
  • Capability inventory: The skill allows for file reading, parsing, and potential database operations based on the processed CSV content.
  • Sanitization: Data validation is limited to type checking and format verification (regex for phone/ABN), which does not address the risk of natural language instruction injection within the data fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 11:48 AM