issue-slice-autopilot

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and parses GitHub issue bodies and PR data via commands like gh issue view and gh pr view, and it rewrites and acts on that user-generated issue content as part of its required workflow (Issue Recon → slice selection → implementation), so untrusted issue/PR text could influence agent actions.