Provides tools for searching arXiv papers, retrieving paper details, and accessing cached summaries via the arXiv public API.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-research-agent-system search arXiv for recent papers on multi-agent systems"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A production-quality multi-agent research system with an arXiv MCP server and LangGraph-based research agents.
MCP Server — Wraps the arXiv public API with SQLite caching (search_papers, get_paper_details, get_cached_summary)
LangGraph Agents — Planner, Researcher, Synthesizer with validation & retry logic
Structured Logging — Full state-machine trace as JSONL
Related MCP server: pubmed-search-mcp
# 1. Copy .env.example and fill in your LLM credentials
cp .env.example .env
# 2. Install dependencies
pip install -e ".[dev]"
# 3. Run the MCP server (stdio transport)
python scripts/run_mcp_server.py
# 4. In another terminal, run a research query
python scripts/run_agent.py --goal "Your research topic here"All settings via .env (see .env.example):
Variable | Description | Default |
| OpenAI-compatible API base URL |
|
| API key | required |
| Model name |
|
| SQLite database path |
|
| Trace log directory |
|
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Persistent memory and knowledge graphs for AI agents. Hybrid search, context checkpoints, and more.
Durable agent-to-agent handoffs and shared scratchpad for multi-agent workflows.
Autonomous research agent that pays every source it cites in USDC on Arc via x402 micropayments.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/hemanth2k6/mcp-research-agent-system'
If you have feedback or need assistance with the MCP directory API, please join our Discord server