Skip to main content
Glama
MCP + CrewAI Agentic Integration

MCP + CrewAI Agentic Integration

by vad-007

๐Ÿค– MCP + CrewAI Agentic Integration ๐Ÿš€

A powerful demonstration of Model Context Protocol (MCP) integrated with CrewAI orchestrations, featuring full observability through AgentOps and high-speed inference via Groq.

Python CrewAI FastMCP AgentOps


๐ŸŒŸ Overview

This project bridges the gap between context-aware tools and autonomous agents. It provides a custom MCP server for real-time external data (Weather, News, Notes) while leveraging CrewAI to orchestrate multi-agent workflows.

๐Ÿ—๏ธ Architecture

  • MCP Layer: A FastMCP server exposing tools for real-time data retrieval.

  • Agentic Layer: CrewAI agents specialized in Market Analysis and Research.

  • Inference Layer: Ultra-fast LLMs (Llama 3.1) hosted on Groq.

  • Observability Layer: AgentOps for tracing, cost management, and debugging.


Related MCP server: MCP Tool Server

โœจ Key Features

๐Ÿ› ๏ธ Custom MCP Server Tools

  • โ˜€๏ธ Weather Engine: Real-time meteorology data via WeatherAPI.

  • ๐Ÿ“ฐ News Intelligence: Global news retrieval via Serper (Google Search API).

  • ๐Ÿ“ Contextual Notes: Locally persistent note management for long-term memory.

  • ๏ฟฝ Auto-Summary: Intelligent summarization of collected context.

๐Ÿ‘ฅ Intelligence Crew

  • ๐Ÿ” Market Researcher: Scours data to identify emerging trends.

  • ๐Ÿ“ˆ Data Analyst: Synthesizes research into actionable market insights.

  • ๐Ÿš€ Sequential Workflow: Fully orchestrated execution path for reliable results.


๐Ÿ› ๏ธ Tech Stack


๐Ÿš€ Getting Started

1. Prerequisites

Ensure you have the following installed:

  • uv (Recommended) or Python 3.13+

  • A valid Groq API Key

  • A valid AgentOps API Key

  • A Serper API Key (for News)

2. Installation

Clone the repository and sync dependencies:

git clone https://github.com/vad-007/MCP_Integration_crewai.git
cd MCP_Integration_crewai
uv sync

3. Configuration

Create a .env file in the root directory:

AGENTOPS_API_KEY=your_agentops_key
GROQ_API_KEY=your_groq_key
SERPER_API_KEY=your_serper_key
WEATHER_API_KEY=your_weather_key

4. Running the Project

๐ŸŒ Start the MCP Server

mcp dev main.py

๐Ÿšข Run the CrewAI Integration

python crewai_agentops_integration.py

๐Ÿ” Run Diagnostics

python test_agentops.py

๐Ÿ“Š Observability with AgentOps

This project is fully instrumented. Every run generates a unique replay URL allowed you to:

  • Watch Agent Self-Correction: See exactly how agents reason through tasks.

  • Trace LLM Calls: Monitor every prompt and completion.

  • Analyze Latency: Visualize the execution timeline of your crew.

Check your dashboard at: app.agentops.ai


๐Ÿ“‚ Project Structure

โ”œโ”€โ”€ main.py                    # FastMCP Server implementation
โ”œโ”€โ”€ crewai_agentops_integration.py # Main CrewAI orchestration
โ”œโ”€โ”€ test_agentops.py           # Connectivity & Diagnostic tool
โ”œโ”€โ”€ .env                       # Environment variables (private)
โ”œโ”€โ”€ pyproject.toml             # Project configuration
โ”œโ”€โ”€ uv.lock                    # Dependency lockfile
โ””โ”€โ”€ docs/                      # Troubleshooting & Optimization guides

๐Ÿค Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

  1. Fork the Project

  2. Create your Feature Branch (git checkout -b feature/AmazingFeature)

  3. Commit your Changes (git commit -m 'Add some AmazingFeature')

  4. Push to the Branch (git push origin feature/AmazingFeature)

  5. Open a Pull Request


๐Ÿ›ก๏ธ License

Distributed under the MIT License. See LICENSE for more information.


Developed with โค๏ธ for the AI Community.

Available Tools

4 tools
add_noteB

Append a new note to the note file.

Args: message (str): The note content to be added.

Returns: str: Confirmation message indicating the note was saved.

Inputs: message (required). Returns: result

fetch_weatherB

Fetch current weather for a city

Inputs: city (required). Returns: result

read_notesB

Read and return all notes from the note file.

Returns: str: All notes as a single string separated by line breaks. If no notes exist, a default message is returned.

Inputs: no named parameters. Returns: result

search_newsB

Fetch search results from Google News via Serper

Inputs: query (required). Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelogโ€ฆ

    TDQS

    B3.3/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: add_note and read_notes handle note management, fetch_weather retrieves weather data, and search_news fetches news results. There is no overlap in functionality, making tool selection straightforward for an agent.

    Naming Consistency4/5

    The tool names follow a consistent verb_noun pattern (add_note, fetch_weather, read_notes, search_news), which is predictable and readable. The minor deviation is that 'fetch_weather' and 'search_news' use different verbs ('fetch' vs. 'search'), but the overall pattern remains clear.

    Tool Count3/5

    With only 4 tools, the set feels thin for a server named 'MCP + CrewAI Agentic Integration', which suggests broader agentic capabilities. While the tools cover basic utilities (notes, weather, news), the scope seems limited compared to the implied integration purpose, bordering on under-scoped.

    Completeness2/5

    Inferring the domain as agentic integration utilities, there are significant gaps: no tools for agent coordination, task management, or data processing beyond simple fetches. The note tools lack update/delete operations, and overall coverage is incomplete for enabling complex agent workflows.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      D
      maintenance
      A FastMCP-based server providing tools for weather queries, random number generation, and real-time stock market data retrieval. It enables AI models to interact with external data and perform specific tasks through the Model Context Protocol.
      -
    • F
      license
      Not graded
      quality
      D
      maintenance
      A production-ready Python MCP server providing tools for fetching live weather data, querying local SQLite databases, reading files, summarizing webpages, and performing safe mathematical calculations. It enables MCP-compatible LLM clients to execute these tasks autonomously as part of agentic workflows.
      -
    • F
      license
      Not graded
      quality
      D
      maintenance
      Multi-agent AI tools for email automation (Outlook), weather (OpenWeather API), sticky notes, and arithmetic, all served via FastMCP.
      -
    • F
      license
      Not graded
      quality
      D
      maintenance
      A FastMCP server that exposes a locally-hosted llama.cpp LLM as MCP tools, plus utilities for weather, news, web fetching, file I/O, and more. Enables MCP-compatible clients to use a local GGUF model for text generation without external APIs.
      1
      -