webex

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

• Third-party content exposure detected (high risk: 0.70). The required workflow uses ClawLink to “connect” to Webex and then the agent runs actions/describe/run commands, which at runtime will fetch and ingest Webex-provided free-form text (e.g., message content, action descriptions, or other user-generated Webex data) into the agent’s LLM context via the CLI’s API responses—an outsider-authored source.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs running "npx @useclawlink/cli", which at runtime downloads and executes remote code from the npm package (https://www.npmjs.com/package/@useclawlink/cli), a required external dependency that can directly control the agent's actions.