pr-review-companion

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly instructs the agent to fetch and read GitHub PRs, diffs, commits, and related Issues using commands like "gh pr view " and "gh pr diff ", meaning the agent ingests untrusted, user-generated third‑party content from GitHub and uses it to drive review decisions and next actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches GitHub PR data at runtime (via commands like "gh pr view" and "gh pr diff") and injects the PR content (e.g., https://github.com/org/repo/pull/42) into the agent's context to drive its prompts, so this external URL/content directly controls the agent's instructions and is a required dependency.