lcx-report-bug

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow reads the user’s bug report at step 1 (“Read the user's bug report...”), and that is outsider-authored free text ingested into the investigation before routing.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly clones remote GitHub repositories at runtime (e.g., sync_latest_source calls cloning of "code-yeongyu/lazycodex" and "openai/codex" via "https://github.com/$REPO" / "https://github.com/$REPO.git"), and those fetched sources are used to decide routing and craft issue/PR content, so external repo content can directly control the agent's outputs.