github-do-issue

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow fetches outsider-authored GitHub issue content (issue body and especially comments) via gh issue view ... --json ... comments, which would be read into the agent context for requirement extraction and summarization, creating an indirect prompt-injection channel from arbitrary user-generated text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill parses and uses GitHub issue URLs (e.g. https://github.com/owner/repo/issues/42) and runs gh issue view at runtime to fetch the issue body and comments, which are injected into the agent's context and directly control implementation instructions.