csv-to-json-validator

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). In the required runtime workflow (parse/load/validate), the tool reads the user-supplied CSV cells (and schema from a provided YAML/JSON source) which are outsider-authored content that can contain free text, and these values are then ingested into the LLM context as part of validation/error reporting (e.g., emitting per-row message strings and errors derived from cell content).