pull-merge-request-creation

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly instructs using MCP tools to "Search related issues and PRs (GitHub) / MRs (GitLab)" and its "External Content Handling" section treats fetched issues, PRs, discussions, and external URLs as inputs, which means the agent will fetch and read untrusted, user-generated third‑party content as part of its required process (SKILL.md Step 1 and External Content Handling).