capsule-crm

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 Capsule CRM and then the agent runs actions/describe/run, which at runtime will fetch Capsule CRM data (e.g., records, notes, and other user-entered text) and feed it into the agent’s LLM context; this content is authored by Capsule CRM users other than the operating user, making it outsider free text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs running "npx @useclawlink/cli" which fetches and executes remote code from the npm package (https://www.npmjs.com/package/@useclawlink/cli), so this external dependency runs code at runtime and can directly control agent actions.