glue-mcp
A model context protocol (MCP) server to AWS Glue Data Catalog
Documentation
glue-mcp
A model context protocol (MCP) server to AWS Glue Data Catalog
Usage
Once you've built a binary (see below), ensure you have your AWS credentials set.
Then simply start the server:
./glue-mcp
2025-04-29T21:25:37.623408Z INFO glue_mcp: Metrics & logging initialised
2025-04-29T21:25:37.623453Z INFO glue_mcp::util: Starting server on 127.0.0.1:8000The server is ready for MCP calls on port 8000.
Development
Nix
Just run `nix develop`, then `cargo build`.
Demo
We'll demonstrate using Ollama.
1. Install & configure `mcphost`
go install github.com/mark3labs/mcphost@latestConfigure mcphost with this MCP server:
{
"mcpServers": {
"aws_glue": {
"url": "http://localhost:8000/sse",
"transport": "sse"
}
}
}2. Configure Ollama
Get a model:
ollama pull llama3.1:latest4. Run `glue-mcp`
cargo run5. Run mcphost & query
~/go/bin/mcphost --config ./mcp.json --model ollama:llama3.1
2025/04/27 11:46:02 INFO Initializing server... name=aws_glue
2025/04/27 11:46:02 INFO Server connected name=aws_glue
2025/04/27 11:46:02 INFO Tools loaded server=aws_glue count=3
You: list databases
2025/04/27 11:47:11 INFO ๐ง Using tool name=aws_glue__list_databases
Assistant:
The available databases are 'reference_data' and 'unstructured'.
You: list tables per database
2025/04/27 11:47:40 INFO ๐ง Using tool name=aws_glue__get_database_metadata
2025/04/27 11:47:45 INFO ๐ง Using tool name=aws_glue__get_database_metadata
Assistant:
The 'unstructured' database is currently empty of tables. The 'reference_data' database also does not contain any
tables.Frequently asked questions
What is glue-mcp?
glue-mcp is A model context protocol (MCP) server to AWS Glue Data Catalog
How do I install glue-mcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is glue-mcp open source?
Yes โ it is hosted on GitHub at https://github.com/atcol/glue-mcp and has 1 stars.
Related MCP tools
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Search infrastructure for AI
An open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM for the Model Context Protocol. Enhance AI
YC (S26) | Open Computer History | Record your screen continuously locally and provide context to your agents (Claude, Codex, Openclaw, Hermes, Runner...)
The fastest and the most accurate file search SDK for AI agents, Neovim, Rust, C, Python, Bun and NodeJS
Semantic version control => entity-level diffs, blame, and impact analysis on top of git. 28 languages via tree-sitter. Built for coding agents.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP