cavalry

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The package/docs describe and ship an explicit "Cavalry Bridge" that will execute arbitrary JavaScript POSTed to 127.0.0.1:8731, and the API surface (api.exec, WebServer/WebClient, runProcess, file I/O, system-info calls) provides clear primitives that enable remote code execution and data exfiltration if the bridge or a script is abused.

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

• Third-party content exposure detected (medium risk: 0.30). The Cavalry Bridge control path ingests arbitrary text content that the user (outsider) POSTs to the local WebServer (127.0.0.1:8731) and makes it available to the running bridge script for execution without requiring the agent to first select a specific item.