github-triage

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md Step 1 ("Gather context") explicitly instructs the agent to read the full GitHub issue body and all comments (user-generated content on GitHub), which are untrusted third-party inputs that the agent uses to decide labels, post comments, and drive follow-up actions—enabling indirect prompt injection.