Drive OriginLab Origin scientific plotting from AI chat via MCP: 62 tools. Modal-dialog watchdog, LabTalk safety gate, release/reconnect, fit with initial/fixed/weighted params, fine-grained plot and sheet edits, matrix tools, FigureSpec declarative YAML (diff-able and replayable), MCP Resources read-only snapshot, matplotlib-figure bridge, PowerPoint assembly with panel labels, Graph Gallery template search, EPS export, file-access whitelist, visual regression baseline, and editable OPJU delivery with deterministic readback verification.
Install
# from npm (prebuilt)
dsh plugin --profile web add dsh-origin-plugin
# from a prebuilt release tarball
dsh plugin --profile web add "https://github.com/Fantasality/dsh-origin-plugin/releases/download/v2.6.1/dsh-origin-plugin-2.6.1.tar.gz"
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:Fantasality/dsh-origin-plugin
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).
README
New here? Start with the Beginner's Quickstart — four ways to run it (Origin button / copy-paste scripts / MCP client / inside DSH), pick yours, done in 1–3 minutes. This page is the capability reference; the Chinese README.md is authoritative.
Drive OriginLab Origin from AI chat over MCP — on your own machine, over COM, with no network round-trip to anyone else's server.
Why this instead of a text-to-figure tool
The deliverable is an editable Origin project file (OPJU) plus a reproducible FigureSpec YAML — every line, marker and axis property stays hand-tunable in Origin and reproducible months later. Text-to-figure / text-to-SVG routes (AutoFigure-like) emit one-off vector objects aimed at method diagrams, not experimental data plots; editability granularity and scientific reproducibility are on different levels.
Install
| Route | Command | Best for |
|---|---|---|
| A. Origin button | python scripts/build_origin_app.py, then install the generated .opx |
People who never touch a terminal |
| B. Copy-paste scripts | Send skills/origin-scripting/SKILL.md to any AI |
Zero install, works with any AI |
| C. MCP client | python install.py (auto-detects Cursor / Claude Desktop / Kimi / Cline / Continue / VS Code) |
AI client users |
| D. HTTP (shared Origin) | python origin_mcp_http.py --port 8731 |
Several AIs, one Origin |
| E. Inside DSH | Install dsh-origin from the plugin market |
DSH users |
| F. npx | npx dsh-origin-plugin |
Quick try |
Requirements: Windows, Origin 2021+ (2026 tested), Python 3.10+.
Fast path: one call, one figure
{"columns": {"x":[1,2,3,4,5], "y":[1,4,9,16,25]}, "intent": "journal", "fmt": "png"}
origin_figure does import → plot → verify → export → (optional) delivery in a single call
(~1 s). The old route took 6–10 tool round-trips; because ~80% of perceived latency is model
decision turns, collapsing the turns is what actually makes it feel fast.
What's in the box (70 tools / 34 error codes)
- Plotting — 2D (line/scatter/line-symbol/column/histogram/box/bar/error bars), 3D, contour, domain templates (stacked spectra, XRD triple, dual-Y, forest, multi-panel)
- Fine-grained editing — per-curve color/width/symbol/visibility, axis, legend anchors, page geometry, window management, remove/swap curves, sort/transpose
- Analysis — fitting with initial/fixed/weighted params, peak analysis, statistics batch (t-test / ANOVA / PCA / Kaplan-Meier), FFT, integration, correlation
- Matrix tools — write / read / plot matrices (surface, contour, wireframe)
- Bridges — matplotlib figure → Origin, PowerPoint assembly with panel letters, OriginLab Graph Gallery template search, EPS/SVG/PDF/TIF/EMF export
- Reusable style templates — save a finished graph as a template and apply it to others (lab-wide consistent styling)
- Trust layer — readback graded
verified/readback_only/unverified, modal-dialog watchdog, LabTalk destructive-command gate, export file-magic validation, 12-image dHash visual regression baseline, stable error codes with a recovery map - Declarative — FigureSpec YAML (diff-able, replayable), MCP Resources read-only snapshots
Science boundaries
No fabricated data · uncertain columns raise a confirmation question · derived columns are labelled · statistics carry a confidence note · unsupported operations are refused explicitly rather than silently ignored. 15 measured failure modes are documented in COMPATIBILITY.md.
Test
python -m pytest tests/ -q # offline tests always; live tests skip without Origin
python origin_mcp_server.py --offline-test
python origin_mcp_server.py --selftest # full chain, needs Origin
MIT License. Not affiliated with OriginLab.
Links
More in this category
Tencent/WeKnora#dsh-weknora★ 37587
Four read-only tools over a WeKnora knowledge base: list knowledge bases, hybrid passage search, reassemble one document's chunks in order, and WeKnora's own cited RAG or ReAct-agent answer with a resumable session id.
DayuanJiang/next-ai-draw-io#dsh-plugin★ 37004
Draws draw.io diagrams in DeepSeek Harness through the next-ai-drawio MCP server: the link in the reply opens a live preview in the right sidebar, you can edit the diagram by hand, and each AI change is a version you can restore. Node.js 18+; npx downloads the MCP server on first use.
superdesigndev/treg★ 8181
Tool catalog for agents: search ~2,600 external endpoints (SEO and SERP, backlinks, social, people and company enrichment, ad libraries, scraping) by the task you want done, read each one's parameters and per-call price, then call it with the credential injected server-side. Ships the skill plus an MCP row that stays disabled until TREG_TOKEN is set.
T-Auto/dsh-ops★ 5867
Provides bash, PowerShell 7, and high-performance Rust tools for DSH on Windows to reduce token usage.
sixtysevenlf/dsh-blender-plugin★ 2153
Lets a DSH agent drive a local Blender over a direct TCP channel with 15 tools: viewport frames, custom-angle renders, Python execution, headless and job offload, transaction rollback, and a planning layer of 228 ops across 34 families.
helloHub1212/dsh-windows-mcp★ 1490
Bridges the CursorTouch/Windows-MCP desktop automation server into DSH through the official MCP client bridge, registering its screenshot, UI tree snapshot, mouse and keyboard input, clipboard, process and registry tools as mcp__windows__* tools for the model, and adds a windows_mcp_doctor diagnostic tool. Computer Use is off by default and is switched on from the plugin panel.
Community comments
Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.