github-issue-pr-draft

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md Steps 1–3 and Step 2 "Task-Specific Context Gathering") explicitly reads upstream repository content — e.g., repo issue/PR bodies and comments, templates in .github/ISSUE_TEMPLATE* and PULL_REQUEST_TEMPLATE*, and performs keyword/duplicate searches and commit/diff summaries — which are untrusted, user-generated GitHub content that the agent both ingests and uses to decide drafting and follow-up actions.