MC Porter

Warn

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the execution of arbitrary local commands and subprocesses through the mcporter call --stdio flag. This allows the agent to run local scripts (e.g., 'bun run ./server.ts') as part of the MCP tool invocation process.
  • [EXTERNAL_DOWNLOADS]: The skill's metadata specifies the installation of the 'mcporter' package from the Node.js (NPM) registry to provide its core functionality.
  • [REMOTE_CODE_EXECUTION]: By design, the tool allows calling remote MCP servers via URLs, which involves fetching and interacting with external logic at runtime.
  • [DATA_EXFILTRATION]: The skill supports making network requests to arbitrary external endpoints ('https://api.example.com/mcp.fetch') and managing authentication tokens via 'mcporter auth', which could be leveraged to send data to external services.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 28, 2026, 05:57 PM