b2c-custom-job-steps

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

• Third-party content exposure detected (high risk: 0.80). This skill's documentation and required job-step patterns explicitly show fetching and ingesting arbitrary external content (e.g., HTTPClient calls using APIUrl/APIEndpoint in references/TASK-ORIENTED.md and CHUNK-ORIENTED.md, and FTP/SFTP download examples with Host/RemoteFile), so job steps can read untrusted third‑party data that may materially influence processing and actions.