to-issues

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime path explicitly fetches an issue’s full body and comments from the issue tracker when the user passes an issue reference (issue number/URL/path), so outsider-authored free text from other contributors can be ingested into the agent’s LLM context via that fetched issue content.