to-issues

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow can fetch an outsider-authored GitHub issue (issue number/URL/path provided by the user) via gh issue view <number> --comments, ingesting the issue body and comments (public/third-party text) into the agent context.

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

• Potentially malicious external URL detected (high risk: 0.80). This skill invokes the GitHub CLI (e.g., gh issue view ...), which at runtime fetches issue bodies/comments from GitHub (https://api.github.com or https://github.com) and uses that fetched content to drive the agent's prompts and outputs, so it is a runtime external dependency that directly controls instructions.