working-with-graphite

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly instructs the agent to "read the existing PR description and incorporate the changes" and to submit PR descriptions via the gh tool, which requires ingesting user-generated/public GitHub PR content (untrusted third‑party content) as part of its workflow.