develop-story

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly reads and parses user-generated content from third-party trackers (GitHub issues/pr bodies and comments via gh issue view / gh pr view and the tracker-state-poller Explore subagent, e.g. Phase 0a inline resolution and the resume/artifact verification steps) and uses those results to resolve document paths, decide resume actions, verify QA/PR state, and drive next tool invocations, so untrusted external text can materially influence agent behavior.