research-assistant
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
AI Research Assistant using LangChain, LangGraph, LangSmith, RAG and MCP (Model Context Protocol)
Project Description
A research assistant that breaks a topic into subtopics, assigns research to agents, summarizes findings, and compiles a report.
Related MCP server: DeepResearch MCP
Features
Graph-based agent orchestration with LangGraph
Reproducible tracing with LangSmith
Modular agent design for research tasks
Planner Agent: Breaks the topic into subtopics.
Researcher Agent: Gathers info for each subtopic.
Summarizer Agent: Summarizes and organizes into a report.
Cache agent responses using SQLite
Contextual document retrieval using RAG and ChromaDB
Prompt & context management using MCP
Project Structure
.
├── agents/ # LLM agents (e.g. researcher, reviewer)
├── config/ # Configurations
├── db/ # SQLite store
├── graphs/ # LangGraph workflow
├── mcp/ # Model Context Protocol (MCP) implementation
├── nodes/ # LangGraph nodes
│ └── conditions # nodes conditions
├── rag/ # RAG (retrieval-augmented generation) logic
├── state/ # Shared state classes for LangGraph workflows
├── tests/ # LangGraph test
├── .env.example # Sample environment variables
├── .gitignore
├── Makefile # Task runner
├── requirements.txt # Python dependencies
└── README.md Requirements
Python=3.11.11
Virtual environment (recommended)
make(optional)
To run the project
Step 1:
Create and activate a virtual environment (recommended)
python -m venv .venv
source .venv/bin/activate
# On Windows: .venv\Scripts\activate Step 2:
Option 1: Using Makefile
make setupOption 2: Without Makefile
pip install -r requirements.txtStep 3:
Copy the .env.example file and rename the file to .env
Step 4:
Add API keys to .env.
Key | Description | Link to Get Key |
| Used for Together AI model access | |
| Used for LangSmith tracing/debugging | |
| Used for search results in RAG |
Usage
Step 1:
To run the MCP development server
Option 1: Using Makefile
make run-mcpOption 2: Without Makefile
mcp dev mcp/server.pyStep 2:
Visit
http://localhost:5173to the browser.Change the Command to
pythonChange Arguments to
mcp/server.pyClick to Connect and wait for connection
After establishing the connection, click Tools -> List Tools -> research
Then write the research topic and Run Tool
To Test Graph Workflow
make test-graph # with make
python tests/test_graph.py # without makeTool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Fan out deep research across multiple AI providers, synthesize into one unified report.
Iterative research with configurable depth and breadth for complex topics.
Web search, fetch, extract, and research for AI agents. Markdown output + AI-synthesized answers.
Cited deep research for agents: run research, search graded sources. OAuth or API key sign-in.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables iterative deep research by integrating AI agents with search engines, web scraping, and large language models for efficient data gathering and comprehensive reporting.326MIT
- AlicenseDqualityDmaintenanceA powerful research assistant that conducts intelligent, iterative research through web searches, analysis, and comprehensive report generation on any topic.417 npm27Apache 2.0
- AlicenseNot gradedqualityDmaintenanceA LangGraph-powered research agent that performs iterative web searches using Google Search and Gemini models to generate structured reports with citations. It integrates with MCP-compatible clients like Claude and Cursor to enable sophisticated, multi-step AI research workflows.MIT
- AlicenseAqualityDmaintenanceEnables AI agents to perform professional-grade deep research by aggregating real-time data from multiple sources, evaluating source credibility, and generating comprehensive reports.311Apache 2.0