origin-mcp

origin-mcp is a local Model Context Protocol (MCP) server that lets AI assistants control Origin/OriginPro on Windows. It connects through OriginLab's Python automation interface and exposes tools for importing data, editing worksheets, creating and refining graphs, running Origin analyses, exporting figures, and managing the Origin application lifecycle.

The goal is to let an AI model work with your installed Origin environment directly instead of only generating standalone plotting code.

Highlights

• Import CSV, TSV, TXT, DAT, XLS, and Excel data into Origin worksheets.

• Read, write, sort, clear, and export worksheet data.

• Create common 2D, 3D, contour, statistical, polar, ternary, vector, bubble, image, and matrix-based plots.

• Cover the documented Origin Plot Type ID table with direct MCP tools.

• Inspect and refine graph pages, layers, axes, legends, labels, reference lines, plot styles, and publication-style formatting.

• Run common Origin analysis commands, including fitting, smoothing, integration, differentiation, peak finding, and descriptive statistics.

• Read analysis output worksheets back as JSON and normalize fit parameters and metrics where possible.

• Export figures, preview exported images, save projects, and release or close Origin safely.

Requirements

• Windows

• Origin or OriginPro installed and licensed

• Python 3.10 to 3.12 recommended

• Origin's originpro package and pywin32

Python 3.14 may run the MCP server itself, but Origin automation packages may not publish compatible wheels yet. If installation fails, use Python 3.11 or 3.12.

Install

python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install -U pip
python -m pip install -e ".[origin]"

If originpro is already available from your Origin installation:

python -m pip install -e .

MCP Configuration

Example MCP client configuration:

{
  "mcpServers": {
    "origin": {
      "command": "D:\\origin-mcp\\.venv\\Scripts\\origin-mcp.exe",
      "args": []
    }
  }
}

More examples are in docs/mcp-config.md.

Smoke Test

After installation, run a real end-to-end Origin check:

.\.venv\Scripts\python.exe -m origin_mcp.smoke_test

The smoke test imports examples/sample_data.csv, creates and formats a graph, exports a PNG preview, saves an OPJU project, and detaches from Origin. See examples/smoke-test.md for options.

For broader verification:

.\.venv\Scripts\python.exe -m origin_mcp.smoke_test --analysis --gallery

To regression-check documented Origin Plot Type IDs:

.\.venv\Scripts\python.exe -m origin_mcp.plot_matrix --limit 10

Example Prompt

Import D:\origin-mcp\examples\sample_data.csv, plot signal_a and signal_b
against time as lines, set the title to "Sample Signals", and export the graph
to D:\origin-mcp\output\sample_labeled.png.

Documentation

• Tool and compatibility reference

• MCP client configuration

• Smoke test

• Plot matrix regression check

Safety

This server can read local data files, write exported figures/projects, and control a local Origin session. Run it only for trusted MCP clients. Use ORIGIN_MCP_ALLOWED_ROOTS to restrict file access when needed.

If Origin says it is being controlled by another program, call origin_detach first. Use origin_force_quit only after confirming there is no unsaved work.

License

MIT. See LICENSE.