CLIO Hdf5
by iowarpcommunityv2.0.1
24
25
CLIO Kit is a collection of Model Context Protocol (MCP) servers designed to facilitate AI agents' interaction with HPC resources and scientific data formats for research purposes.
From the registry: HDF5 FastMCP - Scientific Data Access for AI Agents | CLIO Kit MCP Server
Install & configure
This server lives in a monorepo. The README covers the whole collection, so we can't extract a single canonical install. clio-kit is a monorepo of 16 MCP servers for scientific computing. Individual servers are launched via 'uvx clio-kit mcp-server <name>' (e.g. hdf5).
Tools (16)
adios-mcp
Read data using ADIOS2 engine
arxiv-mcp
Fetch research papers from ArXiv
chronolog-mcp
Log and retrieve data from ChronoLog
compression-mcp
File compression with gzip
darshan-mcp
I/O performance trace analysis
hdf5-mcp
HPC-optimized scientific data with 27 tools, AI insights, caching, streaming
jarvis-mcp
Data pipeline lifecycle management
lmod-mcp
Environment module management
ndp-mcp
Search and discover datasets across CKAN instances
node-hardware-mcp
System hardware information
pandas-mcp
CSV data loading and filtering
parallel-sort-mcp
Large file sorting
paraview-mcp
Scientific 3D visualization and analysis
parquet-mcp
Read Parquet file columns
plot-mcp
Packages
clio-kit
pypiv2.0.1transport: stdio
Details
- Language
- Python
- License
- BSD-3-Clause
- Released
- Feb 23, 2026
- Last commit
- May 21, 2026
- Registry name
- io.github.iowarp/hdf5-mcp