issue-to-implementation

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

• Third-party content exposure detected (high risk: 1.00). This skill fetches and ingests arbitrary, user-generated GitHub issue bodies and comments (see SKILL.md Phase 1 and scripts/fetch_issue.sh which call gh issue view ... --comments), and those untrusted third‑party contents are explicitly read and used to drive analysis, implementation plans, and code/PR actions.