rocketreach

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

• Third-party content exposure detected (high risk: 0.75). At runtime the skill uses ClawLink to perform RocketReach lookups via browser/OAuth and then ingests RocketReach’s returned page/data (outsider-authored web/app content) into the agent’s LLM context through the CLI action outputs.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to run "npx @useclawlink/cli" (fetching and executing the package from npm at runtime—see https://www.npmjs.com/package/@useclawlink/cli), so it relies on remote code that is fetched and executed during runtime and can directly control agent actions.