sarif-parsing

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

• Third-party content exposure detected (high risk: 0.70). The required workflow is to load and parse a user-provided SARIF file (e.g., results.sarif) and then extract fields like message.text and rule metadata into the agent’s processing context; since SARIF content is outsider-authored when the operating user did not author the scan output, this can introduce free text from that SARIF into the LLM context.