vercel-automation

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs calling endpoints like VERCEL_ADD_ENVIRONMENT_VARIABLE that require a 'value' parameter (for secrets/API keys), meaning the agent would need to accept and include secret values verbatim in its tool calls/output, creating exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.70). The skill's "Monitor and Inspect Deployments" workflow explicitly calls VERCEL_GET_DEPLOYMENT_LOGS and VERCEL_GET_RUNTIME_LOGS (and other Vercel API endpoints) to fetch build/runtime logs and deployment event data from a user's Vercel account — externally-hosted, user-generated/untrusted content that the agent is expected to read and use to guide debugging and subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires adding and using the MCP server at https://rube.app/mcp and instructs to "Always call RUBE_SEARCH_TOOLS first to get current tool schemas," which indicates the agent will fetch runtime tool schemas from that external URL that directly control its tool-invocation behavior and instructions.