gh-pr

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

• Third-party content exposure detected (high risk: 0.95). The skill’s runtime workflow fetches GitHub PR review thread comment bodies via the bundled scripts/gh-review-comments helper (GraphQL reviewThreads → comments.nodes[].body), which are outsider-authored free text from other users on the PR.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runs the bundled helper which uses gh api graphql to fetch PR review threads from the GitHub PR URL (e.g., https://github.com///pull/), and those fetched review comment bodies are injected and used at runtime to drive the agent's planning and code-change actions, so this is a runtime external dependency that directly controls agent instructions.