implement-issue

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

• Third-party content exposure detected (high risk: 0.85). Yes: the required workflow fetches outsider-authored GitHub/GitLab issue content (issue body, comments, and system activity) via the selected forge adapter (e.g., gh issue view ... --json ... / glab issue view ... --comments --system-logs --output json) and then uses that evidence for planning/diagnosis, which becomes LLM-readable context at runtime.