linear

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

• Third-party content exposure detected (medium risk: 0.30). In the required runtime workflow, the agent’s write-side inputs only become LLM-ingested content when the user/agent asks to read Linear data via the tool calls getIssue (includes description) and listIssueComments (includes body), i.e., outsider-authored text from issue/comment records is fetched and returned to the agent.