--- title: "CLIO Kit: scientific tools and workflows | IOWarp" description: "CLIO Kit gives AI agents the scientific tools and workflows a project needs. Part of CLIO, the open stack IOWarp builds for AI-driven science." url: https://iowarp.ai/clio-kit/ --- > CLIO Kit gives AI agents the scientific tools and workflows a project needs. Part of CLIO, the open stack IOWarp builds for AI-driven science. # Give agents real instruments. CLIO Kit is part of CLIO, the open-source stack that IOWarp, a team at the Gnosis Research Center, builds to connect AI agents to scientific data and systems. It matters because an agent is only as useful as the tools it can reach. CLIO Kit brings together MCP servers, skills, agent definitions, hooks, and workflow plugins, packaged so a team takes what a task needs. The [released launcher](https://github.com/iowarp/clio-kit/releases/tag/v2.12.3) downloads only what you select and installs isolated server dependencies on first launch. IOWarp will launch the full CLIO stack at SC26 in Chicago. Browse the servers [Components and workflows](https://toolkit.iowarp.ai/catalogue) [Source](https://github.com/iowarp/clio-kit) **CLIO Kit** · command line MCP 1. Install the launcher ```sh uv tool install 'clio-kit>=2.12.3' ``` 2. Inspect the catalog ```sh clio-kit mcp-servers ``` 3. Start a server ```sh clio-kit mcp-server hdf5 ``` The interface MCP servers provide tools. Skills provide procedures. Workflow plugins bundle the components a task needs. [How MCP works for science](https://iowarp.ai/topics/mcp-for-science/) 01 / The catalog ## Find your tool. - 01 · Data ### [HDF5](https://toolkit.iowarp.ai/docs/mcps/hdf5) Inspect datasets and attributes and compute per-index statistics. - 02 · Data ### [Parquet](https://toolkit.iowarp.ai/docs/mcps/parquet) Read columns, summarize nulls, and run grouped aggregates. - 03 · Data ### [Pandas](https://toolkit.iowarp.ai/docs/mcps/pandas) Load, filter, profile, and aggregate tabular data. - 04 · Data ### [ADIOS](https://toolkit.iowarp.ai/docs/mcps/adios) Read slices and per-step statistics through ADIOS2 engines. - 05 · HPC ### [Slurm](https://toolkit.iowarp.ai/docs/mcps/slurm) Submit and manage cluster jobs. - 06 · HPC ### [Spack](https://toolkit.iowarp.ai/docs/mcps/spack) Discover, install, and locate scientific packages. - 07 · HPC ### [JARVIS](https://toolkit.iowarp.ai/docs/mcps/jarvis) Build and manage scientific pipelines. - 08 · HPC ### [Darshan](https://toolkit.iowarp.ai/docs/mcps/darshan) Inspect I/O performance traces. - 09 · HPC ### [Lmod](https://toolkit.iowarp.ai/docs/mcps/lmod) Manage environment modules. - 10 · HPC ### [Parallel Sort](https://toolkit.iowarp.ai/docs/mcps/parallel_sort) Sort, filter, and summarize large log files in parallel. - 11 · Visual ### [ParaView](https://toolkit.iowarp.ai/docs/mcps/paraview) Visualize and render scientific data. - 12 · Visual ### [Plot](https://toolkit.iowarp.ai/docs/mcps/plot) Turn tabular results into plots. - 13 · Earth ### [Geo](https://toolkit.iowarp.ai/docs/mcps/geo) Inspect, validate, summarize, and map GeoJSON vector layers. - 14 · Earth ### [Terrain](https://toolkit.iowarp.ai/docs/mcps/terrain) Analyze elevation models and point clouds. - 15 · Earth ### [Seismology](https://toolkit.iowarp.ai/docs/mcps/seismology) Inspect SAC waveform files and archives, compute trace statistics, and analyze earthquake sequences. - 16 · Discovery ### [ArXiv](https://toolkit.iowarp.ai/docs/mcps/arxiv) Find and retrieve research papers. - 17 · Discovery ### [Scientific Catalog](https://toolkit.iowarp.ai/docs/mcps/scientific_catalog) Discover operator-owned datasets. - 18 · Discovery ### [NDP](https://toolkit.iowarp.ai/docs/mcps/ndp) Search datasets across CKAN instances. - 19 · Discovery ### [Web](https://toolkit.iowarp.ai/docs/mcps/web) Run search and durable document fetch tasks. - 20 · Systems ### [ChronoLog](https://toolkit.iowarp.ai/docs/mcps/chronolog) Log and retrieve data through ChronoLog. - 21 · Systems ### [Compression](https://toolkit.iowarp.ai/docs/mcps/compression) Compress and decompress files. - 22 · Systems ### [Node Hardware](https://toolkit.iowarp.ai/docs/mcps/node_hardware) Inspect hardware on the local machine or a remote node. 02 / In practice ## Choose components. Install a workflow. Project installs support Codex, Claude Code, OpenCode, Cursor, Antigravity, and VS Code. Scientific agents and Dataset Report hooks have adapters for Codex, Claude Code, and OpenCode. Other clients can explicitly select `--components-only` for skills and MCP configuration; see [component support](https://toolkit.iowarp.ai/docs/clients#component-support). ```sh clio-kit plugin install clio-scientific-io --client codex --project /path/to/project ``` Installs the scientific I/O workflow’s skills and MCP configuration into the specified Codex project. Add `--dry-run` to preview changes. The [client setup guides](https://toolkit.iowarp.ai/docs/clients) explain each adapter; the [selective installation guide](https://toolkit.iowarp.ai/docs/installation) covers release downloads and native Claude plugins. 1. 01 ### Choose Browse [MCPs, skills, agents, and hooks](https://toolkit.iowarp.ai/catalogue), or select a [workflow plugin](https://toolkit.iowarp.ai/docs/plugins) for scientific I/O, HPC operations, analysis, or research. 2. 02 ### Install For an individual skill, use `clio-kit skill install dataset-explore --target.agents/skills` in a project whose client reads that directory. Configure its required MCP servers separately. 3. 03 ### Verify Reload the client and inspect its skills and tools. Run `clio-kit doctor --server hdf5 --connect`, then try a tool call on known data. Scientific backends and cluster access have their own prerequisites. Related ## Where CLIO Kit is at work. - [Application **Agents in the materials workflow** Materials science](https://iowarp.ai/applications/materials-science/) - [Application **Agents on the National Data Platform** Data infrastructure](https://iowarp.ai/applications/national-data-platform/) - [Application **Provenance and integrity for AI-driven science** DOE Genesis · SPOTTER-AI](https://iowarp.ai/applications/genesis-spotter/) - [Application **CLIO on national research systems** National infrastructure](https://iowarp.ai/applications/national-infrastructure/) - [Research direction **Staging data while agents think** Can reasoning time hide I/O time?](https://iowarp.ai/research/directions/#staging) - [Demo **Zero to agent with CLIO Kit** Planned for 2026](https://iowarp.ai/research/demos/#demo-zero-to-agent-with-clio-kit) - [Demo **EarthScope through the National Data Platform, 2026** Planned for 2026](https://iowarp.ai/research/demos/#demo-earthscope-through-the-national-data-platform-2026) - [Demo **The HPC operator** Planned for 2026](https://iowarp.ai/research/demos/#demo-the-hpc-operator) - [Demo **HPC workflow automation** Recorded 2025](https://iowarp.ai/research/demos/#demo-hpc-workflow-automation) - [Demo **EarthScope seismograph analysis** Recorded 2025](https://iowarp.ai/research/demos/#demo-earthscope-seismograph-analysis) - [Demo **LAMMPS simulation data** Recorded 2025](https://iowarp.ai/research/demos/#demo-lammps-simulation-data) - [Demo **ParaView as an agent tool** Recorded 2025](https://iowarp.ai/research/demos/#demo-paraview-as-an-agent-tool) - [Demo **An IOWarp workflow end to end** Recorded 2025](https://iowarp.ai/research/demos/#demo-an-iowarp-workflow-end-to-end) - [Demo **ADIOS2 data through MCP** Recorded 2025](https://iowarp.ai/research/demos/#demo-adios2-data-through-mcp) - [Demo **Reproducible deployment** Recorded 2025](https://iowarp.ai/research/demos/#demo-reproducible-deployment) Next step ## Make your workflow an instrument. [Open the CLIO Kit guides](https://toolkit.iowarp.ai/docs/intro) [CLIO Kit on the platform](https://iowarp.ai/platform/#kit)