Search

More Filters

Show More
Add Listing
Go Back

mcp agent

0
New
Multi-Agent Platforms & Frameworks
Report

Report Abuse

Basic Information

mcp-agent is a lightweight, composable Python framework for building AI agents that use the Model Context Protocol (MCP). It is designed for developers who want to assemble multi-agent and single-agent workflows that call external MCP servers as tools. The project implements patterns from Anthropic's Building Effective Agents and provides a model-agnostic reference for OpenAI's Swarm pattern. It manages MCP server lifecycle, exposes server tools to LLMs, and wraps workflows as AugmentedLLM objects so patterns can be composed and nested. The repository includes examples and integrations for standalone scripts, Streamlit apps, a Claude Desktop server adapter, and reactive notebooks, making it practical to prototype or deploy agent applications that leverage MCP-aware services.

Links

Stars
7009
App URL
https://github.com/lastmile-ai/mcp-agent
Github Repository
https://github.com/lastmile-ai/mcp-agent/blob/main/README.md

Categorization

App Details

Features
Provides core components such as MCPApp for global config, gen_client and MCPConnectionManager for server lifecycle and persistent connections, and MCPAggregator to present multiple servers as one. Defines Agent objects that expose MCP servers as tool calls and AugmentedLLM interfaces that combine LLMs with tools and memory. Implements workflow patterns: Parallel, Router, IntentClassifier, Evaluator-Optimizer, Orchestrator-workers and a model-agnostic Swarm. Supports signaling and human-in-the-loop callbacks, composability of workflows, examples and templates, app config via mcp_agent.config.yaml, secrets via a gitignored secrets file or .env, and utilities for listing tools and attaching provider-specific LLMs.
Use Cases
mcp-agent reduces boilerplate for integrating LLMs with external capabilities by automating MCP client creation and server management, so developers can focus on agent logic instead of connection plumbing. Its workflow primitives implement common production patterns so teams can compose complex behaviors like fan-out/fan-in, planning and evaluation, routing and intent classification, and multi-agent orchestration. Built-in examples demonstrate RAG, email automation, Streamlit UIs, and Swarm-style multi-agent setups. It supports pausing for human input, persistent server connections, and exposing an agent app as an MCP server, which helps when embedding agents into MCP hosts or running standalone services.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

virtual lab

0 (0 review)
listing image

mnemo

0 (0 review)
listing image

EdgeChains

0 (0 review)
listing image

RAG Agents Accelerator

0 (0 review)
listing image

LLM Zero to Hundred

0 (0 review)

Featured Listings

Tags

Please fill the required fields*