testplane-skill

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md's "Workflow: Explore A Web App With @testplane/cli" and examples (e.g., npx @testplane/cli navigate <url> and snapshot) explicitly instruct the agent to navigate to arbitrary URLs and capture/interpret DOM snapshots, meaning it will fetch and read untrusted third‑party web pages whose content can directly influence subsequent actions (selector choices, clicks, test logic).

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

• Potentially malicious external URL detected (high risk: 0.70). The skill invokes @testplane/cli at runtime (e.g. via npx) which will fetch and execute code from the npm registry (see https://registry.npmjs.org/@testplane/cli/-/cli-0.0.0.tgz), and @testplane/cli is a required dependency used by the skill.