Skip to main content
Glama
Origin Pro MCP Server

Origin Pro MCP Server

by youngminsw

Origin Pro MCP Server

An MCP (Model Context Protocol) server that enables AI assistants like Claude to control OriginLab Origin Pro via COM automation. All operations are reflected in Origin's GUI in real-time — you watch as the AI creates worksheets, plots graphs, and styles figures.

What Can It Do?

  • Worksheet Management — Create workbooks, read/write data, import CSV/Excel, export CSV, column formulas, sort, transpose

  • Matrices & 3D — Matrix books, XYZ gridding, 3D surface/scatter, contour, heatmap, image plots

  • Graph Creation — Scatter, line, line+symbol, column, bar, area, pie, histogram, contour plots

  • Graph Layers & Axes — Log scales, dual Y axis, panels, reference lines, text annotations

  • Plot Styling — Colors, symbols, line width, publication-ready formatting in one call

  • Analysis — Curve fitting, FFT, smoothing, integration, differentiation, interpolation, peak finding

  • Statistics — Descriptive stats, two-sample t-test, frequency counts (via the stats and transform tools)

  • Project Management — New/save/load projects, export all graphs (with pixel-size control)

  • LabTalk Scripting — Direct LabTalk execution with destructive/file-overwrite commands blocked

This MCP server is intentionally Windows-runtime-only. The AI agent or MCP client can run from Windows or WSL, but the MCP server process that talks to Origin must be launched with Windows Python and pywin32. WSL/Linux can edit the project and run non-COM unit tests, but cannot directly control Origin COM.

Related MCP server: OriginLab MCP Server

Quick Start

Agentic Setup (read this if you are an AI agent configuring this server)

Copy-paste rules to self-configure the origin-pro MCP with zero human steps:

  1. Environment requirement. This server only works on Windows with a licensed Origin/OriginPro install and Windows Python with pywin32. COM automation cannot run from WSL or Linux Python. If the agent runs in WSL, it must launch the Windows executable (uvx.exe / Windows origin-pro-mcp), never WSL Python.

  2. Preferred install path. Use uvx (zero setup) or pip install origin-pro-mcp. Both install to a Windows-local location and "just work":

    { "mcpServers": { "origin-pro": { "command": "uvx", "args": ["--quiet", "origin-pro-mcp"] } } }

    From WSL, call uvx.exe instead of uvx (see Option A below).

  3. Critical UNC rule. If you point at a local checkout from WSL, the repo MUST live under a Windows drive (e.g. C:\Users\YOU\Origin-Pro-MCP). Never reference it through a \\wsl.localhost\... UNC path: the background daemon (python -m origin_pro_mcp.daemon) cannot resolve a -m module or a working directory over UNC, so it fails silently (Origin never opens). uvx/pip install avoid this entirely.

  4. Discover skills first. After connecting, call the list_skills tool to see bundled skills (e.g. publication-figure) and when to use each, then get_skill(name) to load the full step-by-step instructions before producing figures/plots/analysis. No file copying is needed — skills ship inside the package and are exposed by the server.

1. Prerequisites

  • Windows with a licensed Origin/OriginPro installation that exposes the Automation Server

  • Python 3.10+ (Windows Python, not WSL)

Tested environment: Origin Pro 2020. Other Origin/OriginPro versions may work if they expose compatible COM Automation Server and LabTalk behavior, but they are not verified by this project yet.

2. Install & Configure

Option A: uvx (recommended — zero setup)

No manual install needed. The MCP client launches the server for you. Just add this to your Claude Code MCP settings when Claude Code is running on Windows:

{
  "mcpServers": {
    "origin-pro": {
      "command": "uvx",
      "args": ["--quiet", "origin-pro-mcp"]
    }
  }
}

uvx automatically downloads and runs the server in an isolated environment. Nothing else to install. The --quiet flag keeps first-run dependency messages out of your MCP client logs.

If Claude Code or another MCP client is running inside WSL, launch the same Windows server by calling Windows uvx.exe directly:

{
  "mcpServers": {
    "origin-pro": {
      "command": "uvx.exe",
      "args": ["--quiet", "origin-pro-mcp"]
    }
  }
}

For a local checkout before publishing/installing, point Windows uvx at the Windows path of the repo:

{
  "mcpServers": {
    "origin-pro": {
      "command": "uvx.exe",
      "args": ["--quiet", "--refresh", "--from", "D:\\04.Agent OS\\Origin-Pro-MCP", "origin-pro-mcp"]
    }
  }
}

Keep the command and args as separate JSON array entries. That avoids quoting problems when a Windows path contains spaces. If WSL cannot find uvx.exe, set command to the full WSL path for the Windows executable, for example /mnt/c/Users/YOU/.local/bin/uvx.exe.

WSL users — the package must live on a Windows-local path, not a \\wsl.localhost\... (UNC) path. The server runs a background daemon (python -m origin_pro_mcp.daemon); Windows cannot resolve a -m module or use a working directory over a UNC path, so launching it from a WSL-filesystem checkout fails silently (Origin never opens and a console window flashes). uvx/pip install already install to a Windows-local location, so they are unaffected. If you point at a local checkout, clone it under a Windows drive (e.g. C:\Users\YOU\Origin-Pro-MCP) and reference that path — do not point the MCP config at a repo inside the WSL filesystem.

Option B: pip install from PyPI

pip install origin-pro-mcp

Then configure Claude Code:

{
  "mcpServers": {
    "origin-pro": {
      "command": "origin-pro-mcp"
    }
  }
}

Option C: Clone and run directly

git clone https://github.com/youngminsw/Origin-Pro-MCP.git
cd Origin-Pro-MCP
pip install -e .
{
  "mcpServers": {
    "origin-pro": {
      "command": "origin-pro-mcp"
    }
  }
}

Note: If Claude Code runs in WSL, make sure the uvx or python command points to your Windows Python, not WSL Python. Origin COM only works from Windows.

3. Origin Startup

You do not need to start Python or Origin manually. The MCP client starts origin-pro-mcp, which launches an isolated Origin instance for the session. By default the Origin window is visible, so you watch the agent create worksheets and plot graphs in real time.

Display mode — set the ORIGIN_PRO_MCP_VISIBLE environment variable in your MCP config:

Value

Mode

Use for

1 (default)

Visible — Origin window shown

Watching the agent work interactively

0

Invisible — Origin runs hidden

Headless/batch runs, many concurrent agents, no window pop-ups

{ "mcpServers": { "origin-pro": {
  "command": "origin-pro-mcp",
  "env": { "ORIGIN_PRO_MCP_VISIBLE": "0" }
} } }

Reliability & recovery (advanced env vars)

The background daemon runs one isolated Origin instance per session. These environment variables harden it against a wedged Origin COM call (a synchronous operation that never returns) and against destructive mistakes. Some default off; the wedge/data-safety ones now default on because they are safe (see each row). Set any to off to opt out.

Variable

Default

Effect

ORIGIN_PRO_MCP_DISPATCH_TIMEOUT

90

Soft budget (seconds) for each tool dispatch. If Origin doesn't respond within it, a persistent per-session watchdog is polled for the modal dialog that is blocking it: when one is found, the error names its exact title and says whether it was already auto-dismissed (retry the call) or is waiting for you to close it by hand; if none is found it falls back to the generic "most likely a modal dialog" message. NOTHING is killed at this stage. Set off/0 to disable.

ORIGIN_PRO_MCP_DISPATCH_KILL_GRACE

90

Grace (seconds) AFTER the soft warning before Origin is force-killed as a last resort (so a wedged session can never permanently hold a pool slot). Total time to a hard reset = timeout + grace (default 180s). Set off/0 for no warning phase (legacy: force-kill straight at the soft budget).

ORIGIN_PRO_MCP_DIALOG_AUTODISMISS

on

Every session runs a persistent watchdog that polls (~2s) for modal dialogs owned by its Origin process and records their titles. By default it also auto-dismisses (closes) each one it finds, so a startup or mid-session dialog no longer freezes the session. Set to 0/off/false/no to keep detection and reporting (dispatch-timeout errors still name the dialog) without the daemon closing it — you then close it by hand in the Origin window.

ORIGIN_PRO_MCP_AUTOSAVE

on

Save the project in place (its own file, same name — like pressing Save) before a destructive op (delete graph/plot, column deletion, project load/new, overwriting a populated sheet, or a confirmed destructive run_labtalk), so a bad edit is recoverable by reloading. It never writes a differently-named copy, and never overwrites a real file with an empty/blanked project (N5-safe). Only saves a project that already has a file on disk. Set off to disable autosave entirely.

ORIGIN_PRO_MCP_AUTOSAVE_INTERVAL

300

Also save the project in place every N seconds (proactive autosave), not just before destructive ops. Applies to agent-isolated sessions with a saved project; the Origin you ATTACH to is left to you. off/0 disables periodic autosave (preflight still runs).

ORIGIN_PRO_MCP_AUTOSAVE_REQUIRED

1

When autosave is on and a required preflight in-place save fails, the destructive op is not run and an error is returned. Set 0 to proceed without saving.

ORIGIN_PRO_MCP_REAP_CLOSE

off

Session lifecycle: by default a session ending gracefully (idle / client disconnect) is detached — the session's worker thread stops but your Origin window is left exactly as it was (original save path and unsaved edits intact) so you keep the project. Set 1 to restore the old save-a-recovery-copy-and-close behavior. (A wedged session's Origin is still force-killed — the only way to free a worker stuck in a synchronous COM call.)

ORIGIN_PRO_MCP_SWEEP_ORPHANS

off

By default a restarting daemon does not kill leftover Origin windows (so a restart never destroys a project you kept open). Set 1 to have startup reclaim leftover Origins (orphan cleanup, at the cost of closing kept windows).

ORIGIN_PRO_MCP_NO_SPAWN

off

Set 1 to stop the shim from auto-respawning the daemon. Use it to shut the daemon down from the process manager and keep it stopped — tool calls then return a clear "daemon not running" error instead of relaunching it.

ORIGIN_PRO_MCP_ATTACH

off

Set 1 so this session attaches to the Origin you already have open (the shared ApplicationSI instance) instead of spawning a fresh isolated one — the agent then works on your currently-open project. Only one session can attach (a second falls back to an isolated instance); other agents keep their own isolated Origins. The attached instance is never force-killed by the daemon (it's yours).

Per-call override: run_labtalk(script, confirm=True, timeout=120) bounds that one call even when a longer/shorter budget than ORIGIN_PRO_MCP_DISPATCH_TIMEOUT is needed (and works even when the timeout is off).

Rollback: unset any of these (or set the timeout to off) to return to the prior behavior — no code change or redeploy required.

Session lifecycle & restarts

Each MCP client process gets its own daemon session (its own isolated Origin). Because sessions and the daemon can restart independently, the daemon keeps a small ledger sidecar (sessions.json, next to the private lockfile) recording each session's last Origin PID and project path. When a new session starts, the daemon reads that ledger once and, on the first successful tool response, piggybacks a short one-time [origin-mcp] notice telling the agent what happened — so it continues the work instead of silently rebuilding into an empty window. What you may see:

  • Your MCP client restarted (new session). Your previous Origin window was detached, not closed (see ORIGIN_PRO_MCP_REAP_CLOSE): the notice says it is still open with your project and to save/close it in the GUI before reloading, or just work in the fresh instance.

  • The daemon restarted and your old Origin is gone. The notice says your project is not loaded and to reopen it with load_project.

  • Ghost windows. Leftover Origins from earlier sessions are preserved by default, so they can accumulate. The notice summarizes how many are still open; close them in the GUI once saved, or set ORIGIN_PRO_MCP_SWEEP_ORPHANS=1 so a restarting daemon reclaims them.

  • Attach (ORIGIN_PRO_MCP_ATTACH=1). If you got the user's open Origin, the notice reminds you that autosave and force-recovery are disabled there — save explicitly and avoid destructive ops. If another session already holds the single attach slot, the notice says you got an isolated Origin instead.

Separately, load_project appends a one-line collision warning to its result when the ledger shows another live Origin still holding the same project file (saving from both would clobber it — close the other first). The notice and the warning are advisory strings only; they never block a call, and a missing/corrupt ledger is treated as empty.

4. Use It

Just ask Claude to work with Origin:

"Create a scatter plot from this data: x=[1,2,3,4,5], y=[2.1,4.0,5.9,8.1,10.0]"
"Apply publication styling to Fig1 with axis labels Temperature (K) and Absorbance (a.u.)"
"Fit a Gaussian to the data in Book1"
"Export all graphs to C:\Users\me\figures\"

File paths can be Windows style (C:\Users\me\fig.png) or WSL style (/mnt/c/Users/me/fig.png) — the server converts WSL paths automatically, so agents running in WSL can pass their native paths.

Agent Location vs Server Runtime

The agent does not have to run on Windows. These setups are valid:

  • Windows agent -> Windows origin-pro-mcp server -> Origin Pro

  • WSL agent -> Windows origin-pro-mcp server -> Origin Pro

The unsupported setup is WSL/Linux origin-pro-mcp server -> Origin Pro, because COM is a Windows API.

Version Support

This project is currently verified only with Origin Pro 2020. The implementation uses Origin's COM Automation Server and LabTalk, which exist across multiple Origin releases, so other versions may work. Treat them as unverified until someone runs the test suite and a real graph/export smoke test on that version.

Direct LabTalk Safety

The run_labtalk tool is available by default for styling, analysis, graph tweaks, and other advanced Origin operations. It blocks common destructive or file-writing LabTalk commands such as project reset, delete, save/open, file dialogs, external script execution, and graph export. Use the typed tools for saving, loading, importing, and exporting.

This is an accident-prevention guard, not a security sandbox for untrusted code.

LabTalk Gotchas (Origin 2020, styling-report fixes)

  • One flag per set call. set <ds> -c color(255,0,0) -cf color(255,0,0); (combining flags in ONE command) silently wipes the plot to black; the same applies to -k/-kf/-z combined (can blank the symbol). Send each flag as its own set <ds> -flag val; call.

  • Tick-label offset is layer.<ax>.label.offsetV / .offsetH (vertical for the x axis, horizontal for y), in % of the tick-label font size, positive = toward the axis. The plausible-looking names .offset, .voff, .hoff, .offsetX/.offsetY, .xOffset/.yOffset are all silent no-ops (they "read back" a value but never move the labels). Use set_tick_labels(offset_pct=).

  • Never write layer.x2.majorTicks / layer.y2.majorTicks. Setting it to 0 wipes the number labels on ALL FOUR axes, not just the opposite side. Use layer.<ax>.ticks = 0 (or axis(op="tick", axis="top"/"right", tick_direction="none")) to remove tick marks instead.

  • Units differ between line width and error-bar width. set -w is ~200 units per point (500 = 2.5pt); error bars use -erw <points> / -erwc <cap width> directly in points — do not reuse the -w scale for error bars, and never style them via a bare set -w/-ew. set_plot_style(error_bar_width=, error_cap_width=) handles the units.

  • The active window matters. layer.*, col(), and %C all target whatever window is currently active — pass window=<name> to run_labtalk to activate it first, or prefer the typed tools (graph_name/book_name params never depend on activation state).

  • A freshly created page needs a moment before its FIRST styling/read/ export command — create_graph/add_plot_to_graph/ungroup_plots handle this internally now; a raw run_labtalk sequence right after CreatePage/plotxy may still need its own settle.

  • Symbol shape -k codes (Origin 2020, re-verified live): 1=square, 2=circle, 3=triangle-up, 4=triangle-down, 5=diamond, 6=plus, 7=x/cross, 8=asterisk. Codes 9-12 render as a dash/vertical-bar/literal glyph, not useful marker shapes.

Architecture

Claude Code (WSL or Windows)
    |  stdio (MCP protocol)
    v
MCP Server (Windows Python + win32com)
    |  COM automation
    v
Origin Pro (GUI visible in real-time)

Direct CLI (no MCP client)

Every MCP tool is also a plain command, so the repo alone can drive Origin — no MCP client, no per-task scripts. Same Windows-runtime rule applies (Windows Python + pywin32 + Origin).

# List all tools with their arguments
python -m origin_pro_mcp.cli list

# Call any tool: simple flags...
python -m origin_pro_mcp.cli list_worksheets
python -m origin_pro_mcp.cli apply_publication_style --graph_name Fig1

# ...or --json for values with spaces/paths (recommended for scripting)
python -m origin_pro_mcp.cli apply_publication_style --json '{"graph_name": "Fig1", "x_label": "Temperature (K)"}'
python -m origin_pro_mcp.cli export_graph --json '{"graph_name": "Fig1", "file_path": "/mnt/c/Users/me/fig1.png"}'

After pip install (or via uvx) the same is available as the origin-pro-cli command, e.g. origin-pro-cli list_worksheets.

Running a WSL agent? Invoke Windows Python with the package on the path, for example from the repo's src/ directory:

cd src && /mnt/c/.../python.exe -m origin_pro_mcp.cli list_worksheets

The CLI reflects over the same tool registry as the MCP server, so it always exposes exactly the tools listed below.

Available Tools (45 total)

Several tools are dispatchers: one tool name with an op/kind/method argument that selects the specific action, so a handful of tools cover what used to be many single-purpose ones.

Project Management

Tool

Description

new_project

Create new empty Origin project

save_project

Save project to .opju file

load_project

Open existing .opj/.opju file

export_all_graphs

Export every graph in the project to image files

save_graph_template

Save a graph as a reusable .otpu/.otp template

Note: create_worksheet, create_matrix, create_graph, create_matrix_plot, import_data, and worksheet_to_matrix return a JSON string (not a sentence) with the actual assigned name — Origin may rename on collision, so read "name" from the result rather than assuming the requested name was used.

Return-format convention: tools that create or read structured data (worksheets, graphs, matrices, imports, fits, stats, transforms) return a JSON string to parse; tools that apply styling or a one-off action (the Styling and Advanced sections below, plus most of Graphing) return a plain human-readable status string. A few dispatcher tools (stats, transform) mix both across their op/method choices — each is documented per-tool.

Worksheet Data

Tool

Description

create_worksheet

Create new workbook

set_worksheet_data

Write column data (JSON arrays)

get_worksheet_data

Read worksheet data as JSON (empty cells → null)

import_data

Import a CSV/text or Excel file (format="auto"/"csv"/"excel"); CSV/text import suppresses Origin's auto-generated sparkline mini-graph windows by default (sparklines=False). file_path may also be a directory or glob to batch-import every matching file, each into its own book

export_worksheet

Export a worksheet to CSV/text

list_worksheets

List open workbooks, graphs, and matrices

manage_columns

Add, delete, or edit columns — op="add"/"delete"/"properties"/"formula"

sort_worksheet

Sort rows by a column (asc/desc)

transpose_worksheet

Transpose rows and columns

Matrix

Tool

Description

create_matrix

Create a matrix book

set_matrix_data / get_matrix_data

Write / read a 2D grid

worksheet_to_matrix

Grid scattered XYZ into a matrix (xyz2mat)

create_matrix_plot

Surface (3D), contour, heatmap, or image from a matrix (with a data-linked color scale)

Graphing

Tool

Description

create_graph

Create graph (scatter, line, line+symbol, column, bar, area, pie, histogram, box, contour, 3d_scatter); optional template= builds it from a saved .otpu/.otp (2D XY types only)

add_plot_to_graph

Add another dataset to an existing graph

delete_graph

Delete a graph window

remove_plot

Remove one data plot from a graph, addressed by index (COM DataPlot.Destroy()) — removes only that plot even if the same dataset is plotted more than once

set_error_bars

Attach Y/X error bars to an existing plot from an error column (no duplicate)

set_layer_geometry

Set a layer's panel position/size (left/top/width/height)

add_second_y_axis

Add a right-Y layer and plot on it

add_layer

Add a panel/axis layer (right-y, top-x, inset, independent)

axis

Configure axes — op="labels"/"range"/"scale"/"tick"/"frame"; op="labels" also sets the right/top-axis title on axis="right"/"top"

annotate

Add an annotation — kind="reference_line"/"text"/"line"/"arrow"

colormap

Apply a palette and/or set the Z color-scale range on a colormapped graph

export_graph

Export to an image file — raster (png/jpg/tif/bmp) or vector (pdf/eps/emf, resolution-independent); width= sets an exact pixel width for raster (default ~1200px wide, height follows the graph's aspect ratio)

ungroup_plots

Break a plot group so each curve can be colored independently

Styling

Tool

Description

apply_publication_style

One-call publication formatting (recommended)

set_plot_style

Set color, line width, symbol shape/size, and open/solid marker

set_graph_font

Set font family, size, and optional bold

set_legend

Configure legend text and position

set_tick_labels

Tick-label numeric format (decimal/scientific/engineering), bold, decimal places, axis→label offset (offset_pct, % of font; +toward axis)

Analysis

Tool

Description

curve_fit

Curve fitting: parameters ± std errors, R², SSR, reduced χ²; optional plot_on_graph, x_min/x_max to restrict the fit range, and peaks=N (gauss/lorentz/voigt) for multi-peak deconvolution

list_fitting_functions

Show available fit functions

transform

Numerical transform on an XY curve — method="integrate"/"differentiate"/"smooth"/"interpolate"/"fft"/"find_peaks"; smooth takes smooth_method="savitzky_golay"/"adjacent"/"binomial"

stats

Statistics on worksheet columns — op="column"/"compare_means"/"frequency"

Advanced

Tool

Description

run_labtalk

Execute LabTalk with destructive/file-writing commands blocked; optional capture reads variables back. If a 2+ statement script fails as a whole, it is automatically retried statement-by-statement and each statement's OK/FAILED status is reported (partial application is possible — this is intentional)

get_labtalk_variable

Read a LabTalk variable value

Skills

Tool

Description

list_skills

List bundled skills (name, title, when to use)

get_skill

Load a skill's full markdown instructions by name

Example: Publication-Quality Figure

# 1. Start fresh
new_project()

# 2. Create data
create_worksheet(book_name="Data")
set_worksheet_data(
    book_name="Data", sheet_name="Sheet1",
    columns="[[300,350,400,450,500,550,600],[0.12,0.35,0.89,1.62,1.95,1.92,1.25],[0.08,0.22,0.62,1.35,1.88,1.90,1.12]]",
    column_names="Temperature,Pristine,Annealed"
)

# 3. Create line+symbol graph
create_graph(graph_name="Fig1", data_book="Data", data_sheet="Sheet1",
             x_col=1, y_col=2, plot_type="line+symbol")
add_plot_to_graph(graph_name="Fig1", data_book="Data", data_sheet="Sheet1",
                  x_col=1, y_col=3, plot_type="line+symbol")

# 4. One call does everything: colors, fonts, ticks, frame, legend
apply_publication_style(
    graph_name="Fig1",
    x_label="Temperature (K)",
    y_label="Absorbance (a.u.)",
    x_min=280, x_max=620, y_min=0, y_max=2.2,
    legend_entries="Pristine,Annealed",
    legend_position="top-right"
)

# 5. Export
export_graph(graph_name="Fig1", file_path="C:\\Users\\me\\fig1.png")

Result: A publication-ready figure with bold Arial labels, colorblind-safe colors (blue circles + red triangles), filled symbols, solid lines, inward ticks, closed frame, and positioned legend.

Claude Code Skill: Publication Figure

This server ships a bundled skill (src/origin_pro_mcp/skills/publication-figure.md) that teaches Claude how to create journal-quality figures step by step. No manual copying is needed — the skill is packaged inside the wheel and exposed over MCP, so any connecting agent discovers it automatically:

  1. Call the list_skills tool — it returns each skill's name, title, and when to use it (e.g. publication-figure).

  2. Call get_skill("publication-figure") to load the full markdown instructions.

  3. Each skill is also available as an MCP resource at skill://<name> (e.g. skill://publication-figure) for clients that browse resources.

When you ask Claude to "make a publication figure", it can autonomously pull this skill and follow its workflow:

  • Ask about data source, figure type, target journal

  • Use colorblind-safe color palette (steel blue → rose → teal → amber → purple → gray cyan; see the Color Palette table below)

  • Apply proper typography (Arial, bold, correct sizes)

  • Follow a pre-export checklist

The skill also documents COM quirks observed while testing on Origin Pro 2020 — what works, what doesn't, and tested workarounds. This is invaluable if you need to customize beyond apply_publication_style.

Customizing the Skill for Your Style

The included skill is a generic starting template for paper-grade Origin figures. You should copy and customize it to match your lab's habits, target journals, and visual taste:

  • Font: Change from Arial to your journal's preferred font (e.g., Helvetica, Times New Roman)

  • Font sizes: Adjust axis title/tick label/legend sizes to match your journal's figure guidelines

  • Color palette: Replace the default colorblind-safe palette with your group's standard colors

  • Default export path: Set to your working directory

  • Figure recipes: Add templates for your common figure types (XRD patterns, IV curves, etc.)

  • Journal presets: Add specific formatting rules for your target journals (Nature, ACS, RSC, etc.)

Pull the skill with get_skill("publication-figure") (or copy src/origin_pro_mcp/skills/publication-figure.md) into your project and edit freely — it's meant to be a starting point, not a rigid template.

Key Origin Pro 2020 COM Quirks (documented in skill)

Issue

Workaround

Bold axis titles (xb.bold) doesn't exist

Use \b(text) markup in xb.text$

legend.text$ doesn't support multiline via COM

Set column Long Names, then legend -r

legend.x/y uses data coordinates, not %

Calculate from layer.x.from/to

%C notation fails via COM

Use actual plot names from DataPlots

expGraph needs path:=/filename:=/overwrite:=replace (a full path or missing args opens a dialog)

export_graph writes the file directly via expGraph, no clipboard involved

nlr.r2 returns 0 after nlend

Read statistics BEFORE nlend

Combining multiple -flags in ONE set command corrupts the plot (color reset to black, or symbol blanked)

Send one set <ds> -flag val; call per flag, not a delay — see "LabTalk Gotchas" above

layer.x2.majorTicks/layer.y2.majorTicks wipes ALL axes' number labels

Use layer.<ax>.ticks = 0 instead

[Book]Sheet!col(n).type = ... silently ignored

Activate the sheet, then use wks.col(n).type

set <plot> fails when the graph isn't active

Run win -a <graph> before set commands

Typed LabTalk locals (int x = ...) unreadable later

Use untyped assignment to read values back via COM

A graph loaded from a .opju can report zero data plots over COM (per-curve styling/ungrouping silently no-ops)

The core per-curve/axis/frame tools (set_plot_style, ungroup_plots, remove_plot, axis range/scale/tick) now activate the page and re-acquire a fresh layer handle before each call; if the layer is still empty, the tool raises an actionable error instead of returning fake success — recreate the graph in-session if you hit it. Text/font/legend tools (set_graph_font, set_legend) still go through plain LabTalk and can silently no-op on a loaded graph — verify those visually

Dual-Y right axis looks unstylable via layer.y.*

Colour it via layer 2's y2: add_second_y_axis(..., color="r,g,b"), or layer.y2.color (line) + layer.y2.label.color (tick numbers) with the graph active + page.active=2

Heatmap/colormap stuck at ~8 discrete bands

colormap(graph, levels=32..64) (uses numMajorLevels + setLevels(1) + updateScale()); a palette recolors even when the PNG byte-size is unchanged

Hard limitations (no scriptable path on Origin 2020)

Not scriptable

Do this instead

Bar/column FILL PATTERN (hatch) — every route no-ops (set -pfp/-pfw/-pfc, indexed -pfpd/-pfpi, rp.pattern.*, layer.plotN.pattern.*), verified live

Apply hatching in PowerPoint/Illustrator, or distinguish bars by solid color

Plot TRANSPARENCY / alpha — set -paap/-paal/-paas, rr.transparency, layer.transparency, page.transparency all no-op, verified live

Use fully-opaque colors, or composite the alpha in PowerPoint (unconfirmed future candidate: originpro Python Plot.transparency())

Supported Plot Types

Type

Description

Data

scatter

Scatter plot (symbols only)

X, Y

line

Line plot (no symbols)

X, Y

line+symbol

Line with symbols (recommended for publications)

X, Y

column

Vertical bar chart

X, Y

bar

Horizontal bar chart

X, Y

area

Area plot

X, Y

pie

Pie chart

X, Y

histogram

Histogram

single Y column

box

Box chart

single Y column

contour

Filled contour map

X, Y, Z (pass z_col)

3d_scatter

3D scatter (OpenGL)

X, Y, Z (pass z_col)

Matrix plots (surface, contour, heatmap, image) come from a matrix via create_matrix_plot.

Supported Fitting Functions

Category

Functions

Linear

line

Polynomial

poly2, poly3, poly4, poly5

Exponential

exp1, exp2, expgrow1, expdecay1

Peak

gauss, lorentz, voigt

Growth/Sigmoidal

boltzmann, hill, logistic, lognormal

Other

power, sine

curve_fit returns the fitted parameter values with standard errors plus R², SSR, reduced χ², and DoF. Exceptions: power fits and draws the curve, but Origin 2020 does not expose its parameter values over COM, so only the statistics are returned; and a line fit without plot_on_graph uses the fast fitlr path, which returns intercept/slope values and R² only (no standard errors, SSR, χ², or DoF — pass plot_on_graph to get them). Use list_fitting_functions to see the parameter names for each function.

Color Palette

apply_publication_style uses a muted pastel palette (no pure primaries — easier on the eyes, survives grayscale printing, colorblind-distinguishable):

Order

Color

RGB

1st

Soft steel blue

(93, 143, 179)

2nd

Muted rose

(204, 102, 119)

3rd

Muted teal

(68, 170, 153)

4th

Soft amber

(221, 170, 102)

5th

Soft purple

(153, 136, 187)

6th

Gray cyan

(119, 170, 187)

Error bars automatically match their data series color. The fit curve drawn by curve_fit(plot_on_graph=...) uses a muted brick red (170, 68, 80).

Tip: Never use red + green as the only two colors — colorblind users cannot distinguish them.

Troubleshooting

Problem

Solution

"Could not connect to Origin via COM"

Check that Origin/OriginPro is installed and licensed; if it is, run Origin once as administrator to re-register the Automation Server

Tools timeout / "Origin has not responded"

A persistent watchdog polls for modal dialogs on your session's Origin process; by default it auto-dismisses them, and a timeout error names the dialog's exact title and whether it was closed for you. If auto-dismiss is off (ORIGIN_PRO_MCP_DIALOG_AUTODISMISS=0), switch to the Origin window and close/confirm the named dialog yourself — the operation then finishes on its own

"No editable data plots found" on set_plot_style/ungroup_plots/axis ops, or "did not take effect" on an axis-range call

The graph was loaded from a .opju project file in a state Origin freezes over COM (its layer reports zero data plots even after activating the window). This used to silently no-op; it is now a raised error instead of a fake success. Recreate the graph in-session (create_graph/plotxy) or reopen the project fresh — no manual export-and-compare-pixels needed to detect it anymore

"Export failed: ... was not created"

export_graph writes the file directly via expGraph (no clipboard); this means Origin (Windows) could not write to that path — use a Windows path (C:\...) or /mnt/<drive>/... instead of a WSL/Linux path

"... is a Linux/WSL path that Origin ... cannot access"

Any file-path argument (import/export/save/load) given as a bare POSIX path (e.g. /tmp/x.csv) is now rejected upfront with this error rather than silently resolving against the wrong drive — use /mnt/<drive>/... (auto-translated) or a native Windows path

"Window 'X' not found" errors

The error lists every open workbook/graph — use one of those names (Origin may have renamed the window if the name was taken)

Legend missing after styling

Legend uses data coordinates — verify axis range is set before positioning

Symbols appear hollow

Do NOT use set -d flag (it's for dash patterns, not fill)

License

MIT

Available Tools

56 tools
add_arrowA

Draw an arrow from (x1,y1) to (x2,y2) at data coordinates.

The arrowhead sits at the (x2,y2) end (and at the start too when double_headed). Set double_headed=False for a single-ended arrow.

Args: graph_name: Graph name x1, y1: Tail (start) point in data coordinates x2, y2: Head (end) point in data coordinates double_headed: Put an arrowhead on both ends head_size: Arrowhead size in points (default 10)

Returns: Success message

Inputs: graph_name (required), x1 (required), y1 (required), x2 (required), +3 more. Returns: result

add_columnsA

Append empty columns to a worksheet.

Args: book_name: Workbook name sheet_name: Sheet name count: Number of columns to add (default 1)

Returns: Success message

Inputs: book_name (required), sheet_name (required), count. Returns: result

add_layerA

Add a new layer (panel/axis) to a graph.

Args: graph_name: Graph name layer_type: right-y, top-x, inset, or independent

Returns: Success message naming the new layer index

Inputs: graph_name (required), layer_type. Returns: result

add_lineA

Draw a straight line between two data points on a graph.

Useful for guides, connectors, and trend indicators. For a line with an arrowhead, use add_arrow instead.

Args: graph_name: Graph name x1, y1: Start point in data coordinates x2, y2: End point in data coordinates

Returns: Success message

Inputs: graph_name (required), x1 (required), y1 (required), x2 (required), +1 more. Returns: result

add_plot_to_graphA

Add another data series to an existing graph.

Args: graph_name: Existing graph name data_book: Source workbook name data_sheet: Source sheet name x_col: X column number (1-based) y_col: Y column number (1-based) plot_type: Plot type (scatter, line, line+symbol, etc.) y_error_col: Optional Y error column (1-based, 0=none)

Returns: Success message

Inputs: graph_name (required), data_book (required), data_sheet (required), x_col (required), +3 more. Returns: result

add_reference_lineA

Draw a horizontal or vertical reference line at a data value.

Args: graph_name: Graph name orientation: "horizontal" (constant Y) or "vertical" (constant X) value: Axis value where the line is drawn

Returns: Success message

Inputs: graph_name (required), orientation (required), value (required). Returns: result

add_second_y_axisA

Add a right-side Y axis layer and plot a second dataset on it.

Args: graph_name: Existing graph data_book, data_sheet: Source data x_col, y_col: Columns for the right-axis series (1-based) plot_type: scatter, line, line+symbol, column, bar, area

Returns: Success message

Inputs: graph_name (required), data_book (required), data_sheet (required), x_col (required), +2 more. Returns: result

add_text_annotationA

Add a text label to a graph at data coordinates.

Args: graph_name: Graph name text: Annotation text (no quotes, line breaks, or ';') x: X position in data coordinates y: Y position in data coordinates name: Internal object name (letters/numbers/underscore)

Returns: Success message

Inputs: graph_name (required), text (required), x (required), y (required), +1 more. Returns: result

apply_color_mapA

Apply a color palette to a contour/heatmap/surface graph.

Args: graph_name: Graph name (must hold a colormapped plot) palette: Palette name. Bundled perceptually-uniform, colorblind-safe maps (recommended for quantitative data): Viridis, Cividis, Plasma, Inferno, Magma; muted/pastel variants matching a soft figure aesthetic: PastelViridis, PastelCividis. Also accepts built-in Origin .pal names, e.g. Heatmap4ColorBlind, GrayScale, RedWhiteBlue, Fire, Temperature.

Returns: Success message

Inputs: graph_name (required), palette. Returns: result

apply_publication_styleA

Apply complete publication styling to a graph in ONE call.

Sets bold Arial labels, a muted pastel color palette, 2.5 pt lines, readable tick spacing, inward ticks, closed frame, and a borderless bold legend all at once. Designed to minimize token usage — call this once instead of many separate tools.

Args: graph_name: Graph name x_label: X axis label with units, e.g. "Temperature (K)" y_label: Y axis label with units, e.g. "Absorbance (a.u.)" x_min: X axis minimum (None=auto) x_max: X axis maximum (None=auto) y_min: Y axis minimum (None=auto) y_max: Y axis maximum (None=auto) legend_entries: Comma-separated legend entries, e.g. "Sample A,Sample B" legend_position: top-left, top-right, bottom-left, bottom-right

Returns: Summary of applied styling

Inputs: graph_name (required), x_label, y_label, x_min, +5 more. Returns: result

column_statisticsC

Descriptive statistics for one worksheet column.

Returns: JSON: mean, sd, se, variance, median, min, max, sum, n

Inputs: data_book (required), data_sheet (required), col (required). Returns: result

compare_meansB

Two-sample t-test between two columns.

Returns: JSON: t, df, p_value, mean1, mean2, equal_variance

Inputs: data_book (required), data_sheet (required), col1 (required), col2 (required), +1 more. Returns: result

create_graphA

Create a graph from worksheet data.

Args: graph_name: Name for the graph window data_book: Source workbook name data_sheet: Source sheet name x_col: X column number (1-based). Ignored for box/histogram. y_col: Y column number (1-based) plot_type: scatter, line, line+symbol, column, bar, area, pie, box, histogram, contour, 3d_scatter y_error_col: Optional Y error column (1-based, 0=none). XY plots only. z_col: Z column (1-based). REQUIRED for contour and 3d_scatter. title: Optional graph title

Returns: Created graph name (may differ from graph_name if it was taken)

Inputs: graph_name (required), data_book (required), data_sheet (required), x_col (required), +5 more. Returns: result

create_matrixA

Create a new matrix book in Origin.

Matrices back 3D surface, contour, heatmap, and image plots.

Args: book_name: Name for the new matrix book rows: Initial number of rows (default 10) cols: Initial number of columns (default 10)

Returns: Created matrix name (may differ from book_name if it was taken)

Inputs: book_name (required), rows, cols. Returns: result

create_matrix_plotA

Plot a matrix as a surface, contour, heatmap, or image.

Args: matrix_book: Matrix book name (see create_matrix / worksheet_to_matrix) plot_type: surface (3D), contour, heatmap, or image graph_name: Optional name for the new graph z_label: Optional Z label with units (e.g. "Intensity (a.u.)"); sets the matrix long name, which drives both the Z-axis title (3D) and the color-scale title.

Returns: Created graph name

Inputs: matrix_book (required), plot_type, graph_name, z_label. Returns: result

create_worksheetA

Create a new workbook with a worksheet in Origin.

Args: book_name: Name for the new workbook sheet_name: Name for the sheet (default: Sheet1)

Returns: Created workbook name (may differ from book_name if it was taken)

Inputs: book_name (required), sheet_name. Returns: result

curve_fitA

Perform curve fitting on worksheet data.

Args: data_book: Source workbook name data_sheet: Source sheet name x_col: X column number (1-based) y_col: Y column number (1-based) function: Fitting function. Built-in options: line, poly2-5, exp1, exp2, expgrow1, expdecay1, gauss, lorentz, voigt, power, lognormal, logistic, boltzmann, hill, sine y_error_col: Y error column (1-based, 0=none) plot_on_graph: Optional name of an existing graph — the fitted curve is drawn on it as a line (paper style: data symbols + fit line). Also keeps the fit report sheets in the workbook.

Returns: JSON with fitted parameters (value + std_error) and statistics (r_squared, sum_sq_residuals, reduced_chi_sq, dof)

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required), +3 more. Returns: result

delete_columnsA

Delete one or more columns starting at a position.

Args: book_name: Workbook name sheet_name: Sheet name col: First column to delete (1-based) count: How many consecutive columns to delete (default 1)

Returns: Success message

Inputs: book_name (required), sheet_name (required), col (required), count. Returns: result

differentiateC

Compute the derivative dY/dX into a new column.

Returns: Success message naming the new derivative column

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required). Returns: result

export_all_graphsA

Export every graph in the project to image files (one per graph).

Uses the same clipboard-based export as export_graph, so the Windows clipboard contents are replaced during export.

Args: output_dir: Output directory (Windows or WSL style). Created if missing. format: Image format: png, jpg, tif, bmp dpi: Unused (kept for API compatibility; size determined by Origin page) width: Unused (kept for API compatibility) height: Unused (kept for API compatibility)

Returns: Per-graph list of exported files

Inputs: output_dir (required), format, dpi, width, +1 more. Returns: result

export_graphA

Export a graph to an image file.

Uses Origin's CopyPage + clipboard, so the Windows clipboard contents are replaced during export. Image size is determined by the Origin page (width/height/dpi are accepted but have no effect).

Args: graph_name: Graph name to export file_path: Output path (Windows or WSL style, e.g. C:\Users\me\fig1.png or /mnt/c/Users/me/fig1.png). Missing directories are created. format: Image format: png, jpg, tif, bmp. Used as the file extension when file_path has none. width: Unused (kept for API compatibility; size determined by Origin page) height: Unused (kept for API compatibility) dpi: Unused (kept for API compatibility)

Returns: Path to exported file

Inputs: graph_name (required), file_path (required), format, width, +2 more. Returns: result

export_graph_sizedA

Export a graph to an image at a chosen pixel size (expGraph).

Unlike export_graph (clipboard, page-size only), this controls the output pixel width/height directly.

Args: graph_name: Graph to export file_path: Output path (Windows or WSL style) width: Image width in pixels (default 1200) height: Image height in pixels (0 = keep aspect ratio) format: png, jpg, tif, or bmp

Returns: Path and pixel/byte size of the exported file

Inputs: graph_name (required), file_path (required), width, height, +1 more. Returns: result

export_worksheetA

Export a worksheet to a CSV/text file (with column headers).

Args: book_name: Workbook name sheet_name: Sheet name file_path: Output path (Windows or WSL style) delimiter: ",", "tab", ";", "|", or " "

Returns: Path and row/column counts written

Inputs: book_name (required), sheet_name (required), file_path (required), delimiter. Returns: result

fftB

Forward FFT of a signal; outputs a spectrum sheet.

Uses the X column spacing as the sampling interval, so the Frequency column is in real units (1 / X-unit).

Returns: JSON: spectrum sheet and the dominant (peak-amplitude) frequency

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required). Returns: result

find_peaksB

Find peaks in an XY curve (local-maximum method).

Args: direction: positive, negative, or both local_points: neighborhood size for the local-maximum search

Returns: JSON list of {x, y} peak positions

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required), +2 more. Returns: result

frequency_countB

Histogram-style frequency counts for one column.

Args: bin_min: lowest bin start bin_max: highest bin end bin_size: bin width (increment)

Returns: JSON list of {center, end, count, cumulative}

Inputs: data_book (required), data_sheet (required), col (required), bin_min (required), +2 more. Returns: result

get_labtalk_variableA

Get the value of a LabTalk variable.

Gotchas: numeric variables that don't exist read as 0, and variables declared with a type (e.g. int x = 5) are script-local — they vanish when the script ends. Use untyped assignment (x = 5) in run_labtalk if you want to read the value back later.

Args: name: Variable name. Use $ suffix for strings (e.g., 'str$')

Returns: Variable value as string

Inputs: name (required). Returns: result

get_matrix_dataA

Read a matrix back as JSON.

Args: book_name: Matrix book name

Returns: JSON object {"rows": [[...], ...]}; empty cells are null

Inputs: book_name (required). Returns: result

get_worksheet_dataB

Read data from an Origin worksheet.

Args: book_name: Workbook name sheet_name: Sheet name

Returns: JSON object with column data

Inputs: book_name (required), sheet_name (required). Returns: result

import_csv_to_worksheetA

Import a CSV/text file into an Origin worksheet.

Args: file_path: Path to the file (Windows or WSL style, e.g. C:\Users\data.csv or /mnt/c/Users/data.csv) book_name: Optional workbook name. Auto-generated if empty. delimiter: Column delimiter (default: comma)

Returns: Name of the created workbook

Inputs: file_path (required), book_name, delimiter. Returns: result

import_excelA

Import an Excel (.xls/.xlsx) file into a new workbook.

Args: file_path: Path to the Excel file (Windows or WSL style) book_name: Optional name for the resulting workbook

Returns: Name of the workbook the data was imported into

Inputs: file_path (required), book_name. Returns: result

integrateB

Integrate Y over X (area under the curve).

Returns: JSON with the integrated area

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required). Returns: result

interpolateB

Resample an XY curve onto evenly spaced X values.

Args: num_points: number of output points across the X range method: linear, spline, bspline, or akima

Returns: JSON: output sheet plus the resampled X and Y

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required), +2 more. Returns: result

list_fitting_functionsA

List available built-in fitting functions and their parameters.

Returns: JSON of function names grouped by category, with parameter names

Inputs: no named parameters. Returns: result

list_worksheetsA

List all open workbooks (with their sheets), graphs, and matrices.

Returns: JSON object: {"workbooks": [{"name", "sheets"}], "graphs": [names], "matrices": [names]}

Inputs: no named parameters. Returns: result

load_projectA

Open an Origin project file. Replaces the current project.

Args: file_path: Path to a .opj or .opju file (Windows or WSL style)

Returns: Success message

Inputs: file_path (required). Returns: result

new_projectA

Create a new empty Origin project (closes current without saving).

Inputs: no named parameters. Returns: result

run_labtalkA

Execute a LabTalk script in Origin Pro.

Use this for any Origin operation not covered by other tools. LabTalk is Origin's built-in scripting language.

Args: script: LabTalk script to execute

Returns: Success/failure message

Inputs: script (required). Returns: result

save_graph_templateA

Save a graph as a reusable Origin template (.otpu/.otp).

Args: graph_name: Graph to save as a template file_path: Output path (Windows or WSL style). The extension is forced to .otpu (or .otp when old_format=True). old_format: Save the legacy .otp format (loadable in Origin 2017 and earlier) instead of the modern .otpu

Returns: Path to the saved template

Inputs: graph_name (required), file_path (required), old_format. Returns: result

save_projectB

Save the current Origin project.

Args: file_path: Output path for the .opju file (Windows or WSL style). ".opju" is appended when no extension is given. If empty, saves to the project's current location.

Returns: Save confirmation with path

Inputs: file_path. Returns: result

set_axis_labelsB

Set axis labels and title for a graph.

Args: graph_name: Graph name x_label: X axis label y_label: Y axis label title: Graph title

Returns: Success message

Inputs: graph_name (required), x_label, y_label, title. Returns: result

set_axis_rangeB

Set axis range for a graph.

Args: graph_name: Graph name x_min: X axis minimum (None=auto) x_max: X axis maximum (None=auto) y_min: Y axis minimum (None=auto) y_max: Y axis maximum (None=auto)

Returns: Success message

Inputs: graph_name (required), x_min, x_max, y_min, +1 more. Returns: result

set_axis_scaleA

Set an axis to linear or a logarithmic scale.

Args: graph_name: Graph name axis: "x" or "y" scale: linear, log10, ln, or log2

Returns: Success message

Inputs: graph_name (required), axis, scale. Returns: result

set_colormap_levelsA

Set the Z range (color scale levels) of a colormapped graph.

Args: graph_name: Graph name (contour/heatmap/surface) z_min: Minimum Z for the color scale z_max: Maximum Z for the color scale

Returns: Success message

Inputs: graph_name (required), z_min (required), z_max (required). Returns: result

set_column_formulaA

Fill a worksheet column from a formula of other columns.

Args: book_name: Workbook name sheet_name: Sheet name col: Target column (1-based); created if it does not exist formula: LabTalk expression, e.g. "col(1)^2", "col(2)*100/col(3)"

Returns: Success message

Inputs: book_name (required), sheet_name (required), col (required), formula (required). Returns: result

set_column_propertiesA

Set a column's long name, units, comment, and/or designation.

Args: book_name: Workbook name sheet_name: Sheet name col: Column (1-based) long_name: Long name (label row) units: Units (label row) comment: Comment (label row) designation: One of none, x, y, z, yerr, xerr, label (blank = leave)

Returns: Success message

Inputs: book_name (required), sheet_name (required), col (required), long_name, +3 more. Returns: result

set_graph_fontB

Set font for graph elements.

Args: graph_name: Graph name font_name: Font family (e.g., Arial) font_size: Font size in points (default 24) target: "all", "axes", "title", "legend", "tick"

Returns: Success message

Inputs: graph_name (required), font_name, font_size, target. Returns: result

set_legendA

Configure graph legend.

Args: graph_name: Graph name visible: Show or hide legend position: top-left, top-right, bottom-left, bottom-right entries: Comma-separated custom legend entries

Returns: Success message

Inputs: graph_name (required), visible, position, entries. Returns: result

set_matrix_dataB

Write a 2D numeric grid into a matrix.

Args: book_name: Matrix book name data: JSON array of equal-length rows, e.g. [[1,2,3],[4,5,6]]

Returns: Success message

Inputs: book_name (required), data (required). Returns: result

set_plot_styleA

Set line/symbol style for a data plot.

Args: graph_name: Graph name plot_index: Data series index (1-based, order the datasets were added; error-bar plots are not counted) line_width: Line width in points (default 2.5) symbol_size: Symbol size (3-20, default 8) symbol_shape: 0=auto, 1=square, 2=circle, 3=triangle-up, 4=diamond, 5=triangle-down, 6=hexagon color: Color name (black, red, blue, green, orange, purple, cyan, magenta)

Returns: Success message

Inputs: graph_name (required), plot_index, line_width, symbol_size, +2 more. Returns: result

set_tick_styleA

Set tick mark style.

Args: graph_name: Graph name tick_direction: "in", "out", or "both" major_length: Major tick length in points (default 8) minor_count: Number of minor ticks between major ticks (default 4) show_minor: Whether to show minor ticks

Returns: Success message

Inputs: graph_name (required), tick_direction, major_length, minor_count, +1 more. Returns: result

set_worksheet_dataB

Write data to an Origin worksheet.

Args: book_name: Workbook name (e.g., "Book1") sheet_name: Sheet name (e.g., "Sheet1") columns: JSON array of arrays, each inner array is a column of data. Example: [[1,2,3],[4,5,6]] for 2 columns with 3 rows column_names: Optional comma-separated column names (e.g., "X,Y,Error")

Returns: Success message

Inputs: book_name (required), sheet_name (required), columns (required), column_names. Returns: result

smoothB

Smooth a curve into a new column.

Args: method: savitzky_golay (sg), adjacent (moving average), or binomial window: number of points in the smoothing window (odd, default 5)

Returns: Success message naming the new smoothed column

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required), +2 more. Returns: result

sort_worksheetA

Sort all rows of a worksheet by one column.

Args: book_name: Workbook name sheet_name: Sheet name col: Column to sort by (1-based) descending: Sort high-to-low when True

Returns: Success message

Inputs: book_name (required), sheet_name (required), col (required), descending. Returns: result

transpose_worksheetA

Transpose a worksheet (rows become columns).

Args: book_name: Source workbook name sheet_name: Source sheet name output_book: Optional new workbook name for the result; if empty, the source sheet is transposed in place

Returns: Success message

Inputs: book_name (required), sheet_name (required), output_book. Returns: result

worksheet_to_matrixA

Convert XYZ worksheet columns into a matrix by gridding (xyz2mat).

Enables 3D surface / contour / heatmap from scattered XYZ data.

Args: data_book: Source workbook name data_sheet: Source sheet name x_col: X column (1-based) y_col: Y column (1-based) z_col: Z column (1-based) rows: Output matrix rows (default 20) cols: Output matrix columns (default 20) matrix_book: Optional name for the output matrix book

Returns: Name of the created matrix

Inputs: data_book (required), data_sheet (required), x_col (required), y_col (required), +4 more. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 56 tools

    Disambiguation5/5

    Each tool has a clear, distinct purpose. Even tools like add_line and add_arrow are differentiated (arrow vs line), and export_graph variants have explicit argument differences. No overlapping functionality that would confuse an agent.

    Naming Consistency5/5

    All tools follow a consistent verb_noun pattern using snake_case (e.g., add_columns, create_graph, set_axis_labels). There are no mixed conventions like camelCase, and the verbs are in present tense. Minor variations like 'frequency_count' are rare and still follow the noun_verb pattern.

    Tool Count1/5

    With 56 tools, the count far exceeds the recommended 3-15 range and is above the 50+ threshold for extreme mismatch. While the server covers a broad scientific domain, the sheer number of tools likely overwhelms an agent, making selection and learning difficult.

    Completeness5/5

    The tool set provides comprehensive coverage for data import/export, worksheet manipulation, graph creation and styling, statistical analysis, curve fitting, signal processing, and project management. There are no obvious missing operations for the intended scientific plotting and analysis domain.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      Enables AI assistants to control Origin/OriginPro on Windows, including data import, worksheet editing, graphing, analysis, and figure export.
      25
      179 PyPI
      157
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI assistants to control Origin 2025b for scientific plotting via natural language, with support for data import, 52+ chart types, curve fitting, statistics, and export.
      9
      MIT