Jira-loop

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text from Jira is ingested at runtime via acli jira workitem view ... --fields ... description (ADF text converted to plain text) and acli jira workitem comment list ... (comment body), and the loop then uses the newest inbound non-operator comment as the “question” to research and write into the LLM context.