skills-workflows

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

• Third-party content exposure detected (high risk: 0.80). The skill's preprocessing and workflow templates (references/preprocessing.md and workflow-templates.md, e.g. the pr-summary flow) explicitly run commands like !gh pr diff and !gh pr view --comments that ingest user-generated GitHub PR diffs/comments which the agent is expected to read/summarize and which can drive follow-up actions (e.g., update-pr), exposing it to untrusted third-party content that could carry indirect prompt-injection.