issue-review

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches issue title, description and all comments from GitHub/GitLab (Step 1 & Step 2 using gh/glab and API calls), plus linked branch diffs and codebase files, and supplies that untrusted, user-generated content into persona sub-agent prompts to drive analysis and posting decisions—creating a clear vector for indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches issue content at runtime from user-provided issue URLs (e.g., https://github.com/owner/repo/issues/42 and https://gitlab.example.com/group/project/-/issues/42 via gh/glab commands) and injects that content directly into persona sub-agent prompts, so these external URLs directly control the agent's prompts and are required for operation.