glw

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via runtime GitLab work-item content fetched from the user’s configured GitLab instance (e.g., glw search/list/view --json returns outsider-authored issue titles/descriptions/comments, which the agent must read to build previews and reports).