integrating-models

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Setup — gather before starting" section explicitly requires the agent to obtain a reference GitHub repo/link and a runnable inference script (public, user-provided content) and to use that reference implementation for integration and parity-testing, which lets third-party instructions materially influence code changes and next actions.