gh-stack

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflows explicitly fetch and parse live GitHub data (e.g., "gh stack view --json", "gh stack sync" which fetches and syncs PR state, and "gh stack checkout " which pulls stacks from GitHub), so it ingests untrusted, user-generated third-party content from public GitHub repos and uses that data to drive decisions like rebasing, syncing, and checkout.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires installing the GitHub CLI extension with "gh extension install github/gh-stack", which fetches and installs remote code from the github/gh-stack repository (a runtime git URL) that the CLI will execute as an extension, making it a required external dependency that can run remote code.