MCP Catalogs
Home

AgentCrew

by saigontechnology·200·Score 49

AgentCrew is a Python-based multi-agent system that supports MCP for specialized AI collaboration.

ai-llmdeveloper-toolsproductivity
38
Forks
3
Open issues
this month
Last commit
2d ago
Indexed

Overview

AgentCrew provides a platform to create and coordinate multiple specialized AI agents, each with distinct roles such as architect, coder, researcher, or browser operator. The system allows agents to hand off tasks to each other based on their specialties, creating a collaborative workflow environment. It supports multiple AI providers and can be used through desktop GUI, terminal, automated jobs, or HTTP API.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Code review and security analysis in development workflows
you:Research projects with specialized agent roles for information gathering
you:Automation tasks through coordinated agent actions
you:Which AI providers does AgentCrew support?
you:How do I enable MCP tools in AgentCrew?

When to choose this

Choose AgentCrew when you need a complete multi-agent chat application with built-in UI and workflows that also supports MCP for external tool integration.

When NOT to choose this

Don't choose AgentCrew if you need a dedicated MCP server for integration into existing AI tooling, as it's primarily a chat application with MCP support rather than a standalone MCP implementation.

Tools this server exposes

8 tools extracted from the README
  • code_analysis

    Understand repo structure, read files, grep, search

  • file_editing

    Write or modify files with search/replace blocks and backups

  • web_search

    Pull current information from the web

  • browser_automation

    Navigate, click, fill forms, and capture screenshots

  • command_execution

    Run safe shell commands with rate limits and audit logs

  • memory

    Remember past conversations and retrieve relevant context

  • voice

    Speak and listen using ElevenLabs or DeepInfra

  • structured_output

    Enforce JSON schema responses in job mode

Note: Tool names extracted from the 'What Can Agents Do?' section of the README. Descriptions are taken directly from the documentation.

Comparable tools

openai-assistant-mcpclaude-desktopllama-agents

Installation

Installation

**macOS / Linux**

curl -LsSf https://agentcrew.dev/install.sh | bash

**Windows**

powershell -ExecutionPolicy ByPass -c "irm https://agentcrew.dev/install.ps1 | iex"

**pip (any platform)**

pip install agentcrew-ai

Configuration

Add your API key via environment variable:

export ANTHROPIC_API_KEY="sk-ant-..."

Or create a config file at ~/.AgentCrew/config.json

FAQ

Which AI providers does AgentCrew support?
AgentCrew supports Anthropic Claude, OpenAI, Google Gemini, GitHub Copilot, DeepInfra, Together AI, OpenCode, and any OpenAI-compatible endpoint.
How do I enable MCP tools in AgentCrew?
MCP tools are configured through the `mcp_servers.json` file in your ~/.AgentCrew directory. You can edit this file directly or manage it through the GUI settings panel.

Compare AgentCrew with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.