The MCP Company
T

The MCP Company

OpenHands clone project for AI agent evaluation, supporting browser tools, oracle tool sets, and tool retrieval functions
2.5 points
8.4K

What is MCPAgent?

MCPAgent is an intelligent tool retrieval system based on the Model Context Protocol. It allows AI assistants to dynamically discover and use various tools to complete tasks. The system provides tool access interfaces through the MCP server, supporting tool retrieval, invocation, and management.

How to use MCPAgent?

To use MCPAgent, you need to configure the MCP server connection, set LLM parameters, and then guide the AI assistant to use available tools through system prompts. The system supports multiple experimental modes, including browser tools, predefined tool sets, and dynamic tool retrieval.

Applicable Scenarios

MCPAgent is particularly suitable for complex AI tasks that require access to external tools and services, such as web browsing, data querying, file operations, and API calls. It is widely used in intelligent assistants, automated workflows, and tool integration scenarios.

Main Features

Intelligent Tool Retrieval
Dynamically discover and recommend the most suitable tools for the current task based on semantic similarity
MCP Protocol Integration
Fully compatible with the Model Context Protocol standard, supporting multiple MCP servers
Multi-Tool Type Support
Support multiple tool types, such as browser tools, file system tools, and calculation tools
Configurable Experiment Modes
Support different experimental configurations, including browser mode, predefined tool set mode, and dynamic retrieval mode
LLM Integration
Support multiple large language models, with configurable model parameters and authentication information
Advantages
Dynamic tool discovery: No need to know all available tools in advance
Flexible configuration: Support multiple operating modes and experimental settings
Standardized interface: Based on the MCP protocol, with good compatibility
Easy to expand: Can easily add new tools and services
Intelligent recommendation: Semantic-based tool retrieval improves task success rate
Limitations
Dependent on the MCP server: Requires correct configuration and operation of the MCP server
Complex configuration: Requires setting multiple configuration files and parameters
Performance dependence: Tool retrieval and invocation performance are affected by the network and server
Learning curve: Requires understanding of the MCP protocol and related concepts

How to Use

Environment Preparation
Ensure that Docker, Poetry, Python 3.12, and NodeJS 22.x are installed. On the Ubuntu system, you also need to install build-essential, and on WSL, you need to install netcat.
Project Building
Enter the OpenHands directory and run the build command. This process may take some time to complete the installation and configuration of all dependencies.
MCP Server Configuration
Configure the corresponding MCP server according to the experimental requirements. Ensure that the server is running on the correct port (usually port 7879).
LLM Configuration
Configure LLM parameters in the config.toml file, including the model name, authentication token, and other necessary information.
Select Experiment Mode
Select a suitable experiment mode according to the task requirements and configure the corresponding system prompts and tool settings.
Run Evaluation
Execute the corresponding run script to start the evaluation process. The system will automatically execute tasks according to the configuration.

Usage Examples

Web Content Analysis
Use browser tools to access web pages and analyze content, extracting key information
Data Calculation Task
Find a suitable calculation tool through tool retrieval to perform complex mathematical calculations
File Operation Task
Use file system tools to read, process, and write files
Multi-Tool Collaboration Task
Combine multiple tools to complete complex workflows

Frequently Asked Questions

What should I do if the MCP server connection fails?
How to choose a suitable experiment mode?
What should I pay attention to when configuring the LLM?
What should I do if there are dependency errors during the build process?
How to improve the accuracy of tool retrieval?
What types of tools are supported?

Related Resources

OpenHands GitHub Repository
The project's source code and latest updates
Model Context Protocol Documentation
The official specification and documentation of the MCP protocol
Development Environment Setup Guide
Detailed instructions for configuring the development environment
MCP Server Configuration Guide
Instructions for installing and configuring the MCP server

Installation

Copy the following command to your Client for configuration
Note: Your key is sensitive information, do not share it with anyone.

Alternatives

V
Vestige
Vestige is an AI memory engine based on cognitive science. By implementing 29 neuroscience modules such as prediction error gating, FSRS - 6 spaced repetition, and memory dreaming, it provides long - term memory capabilities for AI. It includes a 3D visualization dashboard and 21 MCP tools, runs completely locally, and does not require the cloud.
Rust
9.2K
4.5 points
M
Moltbrain
MoltBrain is a long-term memory layer plugin designed for OpenClaw, MoltBook, and Claude Code, capable of automatically learning and recalling project context, providing intelligent search, observation recording, analysis statistics, and persistent storage functions.
TypeScript
8.7K
4.5 points
B
Bm.md
A feature-rich Markdown typesetting tool that supports multiple style themes and platform adaptation, providing real-time editing preview, image export, and API integration capabilities
TypeScript
14.6K
5 points
S
Security Detections MCP
Security Detections MCP is a server based on the Model Context Protocol that allows LLMs to query a unified security detection rule database covering Sigma, Splunk ESCU, Elastic, and KQL formats. The latest version 3.0 is upgraded to an autonomous detection engineering platform that can automatically extract TTPs from threat intelligence, analyze coverage gaps, generate SIEM-native format detection rules, run tests, and verify. The project includes over 71 tools, 11 pre-built workflow prompts, and a knowledge graph system, supporting multiple SIEM platforms.
TypeScript
7.7K
4 points
P
Paperbanana
Python
8.8K
5 points
B
Better Icons
An MCP server and CLI tool that provides search and retrieval of over 200,000 icons, supports more than 150 icon libraries, and helps AI assistants and developers quickly obtain and use icons.
TypeScript
10.4K
4.5 points
A
Assistant Ui
assistant - ui is an open - source TypeScript/React library for quickly building production - grade AI chat interfaces, providing composable UI components, streaming responses, accessibility, etc., and supporting multiple AI backends and models.
TypeScript
8.6K
5 points
A
Apify MCP Server
The Apify MCP Server is a tool based on the Model Context Protocol (MCP) that allows AI assistants to extract data from websites such as social media, search engines, and e-commerce through thousands of ready-to-use crawlers, scrapers, and automation tools (Apify Actors). It supports OAuth and Skyfire proxy payment and can be integrated into MCP clients such as Claude and VS Code through HTTPS endpoints or local stdio.
TypeScript
9.4K
5 points
G
Gitlab MCP Server
Certified
The GitLab MCP server is a project based on the Model Context Protocol that provides a comprehensive toolset for interacting with GitLab accounts, including code review, merge request management, CI/CD configuration, and other functions.
TypeScript
26.7K
4.3 points
D
Duckduckgo MCP Server
Certified
The DuckDuckGo Search MCP Server provides web search and content scraping services for LLMs such as Claude.
Python
79.3K
4.3 points
N
Notion Api MCP
Certified
A Python-based MCP Server that provides advanced to-do list management and content organization functions through the Notion API, enabling seamless integration between AI models and Notion.
Python
24.4K
4.5 points
M
Markdownify MCP
Markdownify is a multi-functional file conversion service that supports converting multiple formats such as PDFs, images, audio, and web page content into Markdown format.
TypeScript
37.7K
5 points
F
Figma Context MCP
Framelink Figma MCP Server is a server that provides access to Figma design data for AI programming tools (such as Cursor). By simplifying the Figma API response, it helps AI more accurately achieve one - click conversion from design to code.
TypeScript
71.0K
4.5 points
U
Unity
Certified
UnityMCP is a Unity editor plugin that implements the Model Context Protocol (MCP), providing seamless integration between Unity and AI assistants, including real - time state monitoring, remote command execution, and log functions.
C#
37.9K
5 points
G
Gmail MCP Server
A Gmail automatic authentication MCP server designed for Claude Desktop, supporting Gmail management through natural language interaction, including complete functions such as sending emails, label management, and batch operations.
TypeScript
23.6K
4.5 points
C
Context7
Context7 MCP is a service that provides real-time, version-specific documentation and code examples for AI programming assistants. It is directly integrated into prompts through the Model Context Protocol to solve the problem of LLMs using outdated information.
TypeScript
106.1K
4.7 points
AIBase
Zhiqi Future, Your AI Solution Think Tank
© 2026AIBase