plan-linear-issue

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

• Third-party content exposure detected (high risk: 0.85). Step 2 “Fetch the issue” and later “Explore the codebase” read outsider-authored free text from Linear issues (issue Description/labels/content and sub-issues) via the get_issue / list_issues runtime workflow, which is then parsed and used to generate the LLM context/output.