meta-issue-to-pr-autopilot

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

• Third-party content exposure detected (high risk: 0.85). The skill’s required workflow (triage an issue, delegate to sub-agent, draft PR description, open PR) necessarily involves reading the body/comments of an existing GitHub issue authored by someone other than the operating user via gh issue view/issue APIs, which is outsider free text ingested into the agent’s LLM context.