Agentic MCP Client

v1.0.0Coding Agentsstable

A standalone agent runner that executes tasks using MCP (Model Context Protocol) tools via Anthropic Claude, AWS BedRock and OpenAI APIs. It enables AI agents to run autonomously in cloud environments and interact with various systems securely.

anthropic-apianthropic-claudeaws-bedrockaws-bedrock-agentsmcp-bridge
Share:
40
Stars
0
Downloads
0
Weekly
0/5

What is Agentic MCP Client?

Agentic MCP Client is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to standalone agent runner that executes tasks using mcp (model context protocol) tools via anthropic claude, aws bedrock and openai apis. it enables ai agents to run autonomously in cloud environments a...

A standalone agent runner that executes tasks using MCP (Model Context Protocol) tools via Anthropic Claude, AWS BedRock and OpenAI APIs. It enables AI agents to run autonomously in cloud environments and interact with various systems securely.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • A standalone agent runner that executes tasks using MCP (Mod

Use Cases

Run autonomous agents using Anthropic Claude, AWS BedRock, or OpenAI.
Execute MCP tools securely in cloud environments.
Interact with various systems through agentic frameworks.
peakmojo

Maintainer

LicenseApache-2.0
Languagepython
Versionv1.0.0
UpdatedSep 28, 2025
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx agentic-mcp-client

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Agentic MCP Client

Agentic MCP Client is a standalone agent runner that executes long-horizon tasks autonomously by combining MCP tool servers with Anthropic Claude, AWS Bedrock, or OpenAI as the reasoning backend. It manages tool selection, multi-step execution, and iteration limits, making it possible to deploy AI agents in cloud environments that interact with file systems, remote machines, and custom MCP tools — all configured through JSON files rather than code. A built-in web dashboard at localhost:3000 lets you monitor agent runs and task progress in real time.

Prerequisites

  • Python 3.10 or later with the uv package manager installed
  • Node.js 18 or later (for the dashboard)
  • An Anthropic API key, AWS Bedrock credentials, or OpenAI API key depending on the backend you choose
  • Docker (optional, for running MCP tool containers such as mcp-remote-macos-use)
1

Clone the repository and install dependencies

Clone the project and let uv resolve and install all Python dependencies into a managed virtual environment.

git clone https://github.com/peakmojo/agentic-mcp-client.git
cd agentic-mcp-client
uv sync
2

Create the agent configuration file

Create config.json in the project root and set your chosen inference backend credentials. Use base_url and api_key for Anthropic/OpenAI, or set use_bedrock: true with AWS credentials for Bedrock.

{
  "base_url": "https://api.anthropic.com",
  "api_key": "sk-ant-your-anthropic-key",
  "use_bedrock": false
}
3

Define a task for the agent

Create agent_worker_task.json describing the task the agent should complete, the model to use, and iteration limits.

{
  "task": "Find all image files in the current directory and report their sizes",
  "model": "claude-3-7-sonnet-20250219",
  "system_prompt": "You are a helpful agent with access to file system tools.",
  "verbose": true,
  "max_iterations": 20
}
4

Start the monitoring dashboard (optional)

In a separate terminal, install and launch the Next.js dashboard so you can observe agent execution in a browser.

cd dashboard
npm install
npm run dev
5

Run the agent

Execute the agent worker. It will load tools from the configured MCP servers, plan steps, and iterate until the task is complete or max_iterations is reached.

uv run agentic_mcp_client/agent_worker/run.py

Agentic MCP Client Examples

Client configuration

config.json for using Anthropic Claude as the agent backend.

{
  "base_url": "https://api.anthropic.com",
  "api_key": "sk-ant-api03-your-key-here",
  "use_bedrock": false,
  "aws_region": "",
  "aws_access_key_id": "",
  "aws_secret_access_key": ""
}

Prompts to try

Example task definitions for the agent_worker_task.json file.

- "Scan the /Users/me/projects directory and list all Python files larger than 100KB"
- "Fetch the content of https://example.com/api/data and save a summary to summary.txt"
- "Check available disk space on all mounted volumes and report any that are over 80% full"
- "Find all TODO comments in .py files under ./src and compile them into a markdown checklist"
- "Run the test suite in ./tests and report which tests failed and why"

Troubleshooting Agentic MCP Client

uv sync fails with dependency resolution errors

Ensure you are using Python 3.10 or later. Run uv python install 3.12 to install a compatible version, then retry uv sync. If you have conflicting system packages, use uv venv --python 3.12 to create an isolated environment first.

Agent loops indefinitely without completing the task

Lower the max_iterations value in agent_worker_task.json and refine the task description to be more specific. Enable verbose: true to see which tools are being called so you can identify where the agent is getting stuck.

AWS Bedrock authentication fails

Set use_bedrock: true and provide valid aws_region, aws_access_key_id, and aws_secret_access_key values in config.json. Ensure the IAM user or role has bedrock:InvokeModel permission for the model you specified.

Frequently Asked Questions about Agentic MCP Client

What is Agentic MCP Client?

Agentic MCP Client is a Model Context Protocol (MCP) server that standalone agent runner that executes tasks using mcp (model context protocol) tools via anthropic claude, aws bedrock and openai apis. it enables ai agents to run autonomously in cloud environments and interact with various systems securely. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Agentic MCP Client?

Follow the installation instructions on the Agentic MCP Client GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Agentic MCP Client?

Agentic MCP Client works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Agentic MCP Client free to use?

Yes, Agentic MCP Client is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "agentic-mcp-client": { "command": "npx", "args": ["-y", "agentic-mcp-client"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Agentic MCP Client?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides