cursor-subagents

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill follows secure secret management practices in its examples (references/agents-field.md) by utilizing environment variables (process.env.CURSOR_API_KEY) rather than hardcoded credentials.
  • [SAFE]: Documentation in references/mcp-scoping.md explicitly promotes security through 'MCP scoping,' which restricts subagent access to only specific tools or servers required for their tasks, minimizing the risk surface of delegation.
  • [SAFE]: External references to the @modelcontextprotocol/server-github package are legitimate and point to official services from the Model Context Protocol organization.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 08:28 AM