visio-mcp-sci
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@visio-mcp-sciDraw a scientific diagram with blocks and arrows using SCI compact style"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Visio MCP SCI
A Windows MCP server for creating fully editable Microsoft Visio diagrams with compact SCI/IEEE styling, native arrowheads, obstacle-aware routing, and a pre-export geometry audit.

Why this fork exists
General-purpose Visio automation can produce valid diagrams that still have publication problems: touching controls hide arrowheads, hand-routed polylines cross text, decision diamonds become too tall, and dense figures use inconsistent colors or typography. Visio MCP SCI turns those concerns into reusable defaults and executable checks.
Related MCP server: AI Charts
Highlights
56 MCP tools covering documents, pages, primitives, stencils, connectors, alignment, export, and diagram inspection.
sci_compactprofile with print-friendly, low-saturation semantic colors.Flat editable diamonds (
1.20 × 0.54 inby default).Arial 9.5 pt body text and 10 pt multi-line text with 115% line spacing.
Native Visio
BeginArrow/EndArrowproperties—no triangle helper shapes.Automatic minimum clearance before AutoConnect, preventing hidden arrowheads.
Obstacle-aware orthogonal routing with perpendicular terminal segments.
Side-center attachment zones that exclude rounded corners.
A 0.12-inch straight terminal stub before the first and final 90-degree turn.
Layout audit for control overlap, insufficient gaps, short connectors, connector crossings, corner attachments, diagonal terminals, and premature bends.
Strict scientific export that refuses to export a figure that fails the audit.
Semantic roles stored as Visio Shape Data, so validation survives save/reopen.
Requirements
Windows 10 or 11
Desktop Microsoft Visio
Python 3.10 or newer
The currently verified local runtime is Python 3.12.12.
Install
git clone https://github.com/2924547961/visio-mcp-sci.git
cd visio-mcp-sci
python -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install -e .MCP configuration
Codex config.toml:
[mcp_servers.visio]
command = "C:\\path\\to\\visio-mcp-sci\\.venv\\Scripts\\python.exe"
args = ["-m", "visio_mcp"]
startup_timeout_sec = 120JSON-based MCP clients:
{
"mcpServers": {
"visio": {
"command": "C:\\path\\to\\visio-mcp-sci\\.venv\\Scripts\\python.exe",
"args": ["-m", "visio_mcp"]
}
}
}Restart the MCP client after changing its configuration.
Safe scientific workflow
Call
get_scientific_figure_standard("sci_compact").Create shapes with
batch_draw_shapes; assignsemantic_rolevalues such asbackbone,exit,decision,auxiliary,highlight,accept, orreject.Connect ordinary neighbors with
batch_connect_shapes. Its defaultsenforce_clearance=trueandenforce_perpendicular=trueautomatically separate touching controls and select right-angle routing when needed.Use
draw_safe_orthogonal_connectorfor branches that pass near other modules.Call
audit_scientific_layoutand fix every reported issue.Use
export_scientific_figurefor final SVG/PNG export. Strict mode blocks export if either errors or warnings remain.
Example obstacle-aware branch:
{
"from_shape_id": 12,
"to_shape_id": 19,
"obstacle_ids": [14, 15, 16],
"preferred_axis": "horizontal",
"clearance": 0.12,
"terminal_stub": 0.12,
"role": "decision_flow"
}Connector geometry invariants
Every publication connector follows these rules:
Attach in the center region of a shape side, outside the rounded-corner zone.
Leave and enter the attached side along its normal (perpendicular) direction.
Keep at least 0.12 in of straight terminal line before any 90-degree bend.
Use a straight connector only when the two attachment points are axis-aligned.
Use native Visio line patterns and arrowheads; never add triangle shapes.
audit_scientific_layout reports violations as
connector_attaches_near_corner, connector_terminal_not_perpendicular, or
connector_terminal_stub_too_short. Strict export blocks all three.
Compact SCI defaults
Property | Default |
Horizontal gap | 0.18 in |
Vertical gap | 0.16 in |
Minimum control/arrow gap | 0.14 in |
Connector clearance | 0.12 in |
Straight terminal stub | 0.12 in |
Rounded-corner exclusion | 0.10 in |
Flat diamond | 1.20 × 0.54 in |
Body / multi-line text | 9.5 / 10 pt |
Multi-line spacing | 115% |
Data flow | charcoal solid, 1.1 pt |
Decision flow | ochre dashed, 1.0 pt |
All shapes, polylines, and connectors remain editable native Visio objects.
Publication-style basis
The defaults follow common publisher requirements: editable vector artwork, standard fonts, minimal whitespace, restrained colors, clear line work, and no decorative effects. See the official Nature figure guide, IEEE graphics guidance, and Elsevier artwork guidance.
Validation
Unit tests that do not launch Visio:
python -m unittest discover -s tests -p "test_*.py"Windows + Visio integration smoke test:
python tests\integration_layout_safety.pyLicense and attribution
MIT licensed. This repository is derived from the MIT-licensed visio-mcp
0.1.2 distribution by yushun; see NOTICE.md and LICENSE.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseBqualityCmaintenanceAn MCP server that generates beautiful Excalidraw architecture diagrams with perfect auto-layout, stateful editing, and architecture-aware component styling.Last updated24137MIT
- Alicense-qualityDmaintenanceAn MCP server that enables AI assistants to programmatically create and manage flowcharts, ERDs, and swimlane diagrams. It provides tools for manipulating diagram structures, performing auto-layouts, and exporting to Mermaid or Markdown formats.Last updated20MIT
- AlicenseAqualityCmaintenanceEnables AI models to generate interactive visuals including charts, diagrams, UI mockups, and SVG graphics from plain text prompts. This Windows-based MCP server serves as an intermediary to bridge AI tools with visual output capabilities.Last updated1421MIT
- AlicenseAqualityDmaintenanceAn MCP server that generates professional infrastructure diagrams using the Python diagrams DSL, with first-class Azure support and GitHub Copilot integration for natural language diagram generation.Last updated4MIT
Related MCP Connectors
Educational MCP server with 17 math/stats tools, visualizations, and persistent workspace
Autopilot MCP server for GEO analyses, reports, content, audits, memories and agents.
The MCP server for Azure DevOps, bringing the power of Azure DevOps directly to your agents.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/2924547961/visio-mcp-sci'
If you have feedback or need assistance with the MCP directory API, please join our Discord server