Repolode
Submit
apify

apify/mcpc

A command-line MCP client that maps MCP operations like tools, resources, prompts and OAuth login to shell commands for debugging and scripting.

View on GitHubWebsite
Stars
1,091
Forks
107
Open issues
31
Last commit
16 hours ago

What is it?

mcpc is a command-line client for the Model Context Protocol that turns every MCP operation into a shell command. Developers can use it to inspect and debug MCP servers, script repeatable MCP workflows in plain shell, or give AI agents access to MCP through a single Bash tool call.

Sessions are named with an @ prefix and can stay connected to several servers at once, over stdio or Streamable HTTP. Authentication covers OAuth 2.1 with PKCE, client ID metadata documents, dynamic client registration, client credentials, and enterprise-managed authorization, with credentials stored in the OS keychain. A --json flag makes output composable with jq and xargs.

The tool also includes a proxy mode that exposes an authenticated session as a local MCP server without revealing the upstream tokens, grep for searching tools across sessions, and experimental x402 payment support on Base.

Key features

  • 01Maps tools, resources, prompts, async tasks and skills to shell commands over stdio and Streamable HTTP.
  • 02Keeps persistent sessions to multiple MCP servers at once.
  • 03Supports OAuth 2.1 with PKCE, CIMD and dynamic client registration, with credentials kept in the OS keychain.
  • 04Outputs JSON via --json for piping into jq and shell scripts.
  • 05Provides a local proxy mode that hides upstream credentials from clients.
  • 06Searches tools across active sessions with grep.

Use cases

  • Debugging and inspecting a remote MCP server from the terminal.
  • Scripting repeatable MCP workflows with shell pipelines and jq.
  • Giving an AI coding agent MCP access through a single Bash tool without exposing credentials.
  • Running CI jobs that authenticate to MCP servers with OAuth client credentials.

Quick start

  1. 1

    Install with Homebrew

    brew install mcpc
  2. 2

    Or install globally with npm

    npm install -g @apify/mcpc
  3. 3

    Log in to a remote MCP server and save OAuth credentials

    mcpc login mcp.apify.com
  4. 4

    Connect to the server with a named session and list its tools

    mcpc connect mcp.apify.com @test

Things to know

  • Relies on the OS keychain; headless or CI systems fall back to a file-based credential store.

In collections

Claude Code tools, skills & pluginsOpen-source AI agent frameworks & appsMCP serversCLI & terminal toolsLogin, SSO & identity servers

Topics

#mcp#cli#ai-agents#model-context-protocol#oauth#shell#command-line#bash#claude#code-mode

FAQ

What is mcpc?
A command-line MCP client that maps MCP operations like tools, resources, prompts and OAuth login to shell commands for debugging and scripting. mcpc is a command-line client for the Model Context Protocol that turns every MCP operation into a shell command. Developers can use it to inspect and debug MCP servers, script repeatable MCP workflows in plain shell, or give AI agents access to MCP through a single Bash tool call.
Is mcpc free to use? What license does it have?
mcpc is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is mcpc still maintained?
Actively maintained. The last commit to mcpc was on Nov 6, 2026.
What can I use mcpc for?
Debugging and inspecting a remote MCP server from the terminal. Scripting repeatable MCP workflows with shell pipelines and jq. Giving an AI coding agent MCP access through a single Bash tool without exposing credentials. Running CI jobs that authenticate to MCP servers with OAuth client credentials.
How do I install mcpc?
Install with Homebrew: brew install mcpc Or install globally with npm: npm install -g @apify/mcpc Log in to a remote MCP server and save OAuth credentials: mcpc login mcp.apify.com Connect to the server with a named session and list its tools: mcpc connect mcp.apify.com @test
Where can I get mcpc?
The source code and installation instructions are on GitHub: https://github.com/apify/mcpc

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects