monte-carlo-manage-mac

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill contains an explicit instruction to route all MCP tool calls through the plugin's bundled server and to avoid the user's configured MCP server, which is consistent with a deliberate backdoor/data-exfiltration pattern that could capture sensitive table/schema/configuration and credentials.

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

• Third-party content exposure detected (medium risk: 0.30). The skill “Edit workflow” reads the user-provided MaC YAML file via a “Read tool” step (Step 1: “Use the Read tool to load the user's file”), so runtime ingestion includes outsider-authored free text entered by the user into that file.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly instructs fetching the runtime schema via curl (https://clidocs.getmontecarlo.com/mac/schema.json), which is retrieved during execution and used as the authoritative schema to drive validation/authoring behavior.