tanstack-start-client-core-start-core-middleware

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

• Third-party content exposure detected (medium risk: 0.30). In the middleware skill’s documented workflow, client-provided sendContext (free text/values from the requester) is accepted as context in the required .server() phase and is explicitly noted as untrusted client input, creating an indirect prompt-injection/data-injection risk if later code uses it before access checks.