What problem does it solve? Interacting with Model Context Protocol servers typically requires writing custom client code, which slows down debugging, scripting, and AI agent integration. This Skill provides a universal CLI client that maps MCP operations to shell commands with persistent sessions and scriptable JSON output. ## Core Features & Use Cases - Persistent Session Management: Connect to multiple MCP servers simultaneously, keep stateful connections alive, and share sessions across agents. - Full MCP Operations: List and call tools, read and subscribe to resources, get prompts, and manage async tasks from the terminal. - OAuth 2.1 Authentication: Log in to remote MCP servers with credentials stored in the OS keychain, with named profiles for staging and production. - Use Case: An AI agent given only a Bash tool can run mcpc connect mcp.apify.com @apify, search tools with mcpc @apify grep "search", and call them with JSON output piped through jq for downstream processing. ## Quick Start Ask the agent to connect to an MCP server with mcpc, search for a relevant tool, and call it with JSON output.