Skip to main content

An MCP server for creating svg animations.

Project description

MCP SVG Animator

An MCP server for creating animated SVG diagrams. You can create a diagram from text input, photos of sketches and YAML specifications.

You don't normally need to use YAML; just tell Claude code what you want, upload photos of sketches, and then iterate until you've got what you want.

You can see some sample inputs and outputs on the Demo Page

File Output Configuration

By default, the MCP server does not allow writing files to your filesystem. To enable file output (SVG, PNG, or video), create a configuration file at ~/.config/mcp-svg-animator/config.yaml:

file_output:
  allowed:
    - path: "~/diagrams"
      types: [svg, png, webm]
    - path: "/tmp/svg-animator/**"
      types: [svg, png]
    - path: "~/projects/**/output"
      types: [svg, png, webm]

Configuration options:

  • path: Directory where files can be written. Supports ~ expansion and glob patterns (* for single level, ** for recursive).
  • types: List of allowed file types (svg, png, webm).

Files in subdirectories of allowed paths are also permitted.

Features

  • Create and iterate on your requirements using natural language and/or rough diagrams
  • Multiple shape types - circle, rectangle, line, text, path, ellipse, group
  • Connection elements - Draw lines between element centers with automatic positioning
  • Text backgrounds - Auto-generate background panels behind text labels
  • Path segments - Build curves with move_to, line_to, cubic_bezier, quadratic_bezier, arc
  • SMIL animations - Animate attributes and transforms
  • Reusable definitions - Define once, use many times
  • Library imports - Share definitions across diagrams
  • Relative positioning - Reference other elements' positions
  • Direct file output - Write SVG directly to file, reducing token usage
  • PNG export - Generate PNG previews of static diagrams
  • Video export - Record animations to .webm video

Installation

pip install mcp-svg-animator

# Install Playwright browsers (for video generation)
playwright install chromium

Usage with Claude Code

Register the MCP server globally:

claude mcp add --scope user svg-animator -- python -m mcp_svg_animator

Then in any Claude Code session, ask Claude code to create animations:

"Create an animated SVG of a red circle that pulses"

See docs/usage.md for more details.

MCP Tools

You don't need to explicitly invoke these. Ask Claude Code to create the animation you describe, and it will invoke the relevant tools.

Tool Description
create_svg_from_yaml Generate SVG from YAML specification. Optional output_path writes SVG to file; optional png_path generates a PNG preview.
create_animation_video Record SVG animation to .webm video

Development

# Clone the repository
git clone https://github.com/romilly/mcp-svg-animator.git
cd mcp-svg-animator

# Create and activate virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt
pip install -r requirements-test.txt

# Install Playwright browsers (for video generation)
playwright install chromium

# Run tests
pytest

# Type checking
pyright src/

Examples

See the examples/ directory for sample YAML files and generated SVGs:

  • demo.yaml - Basic shapes
  • stick_figure.yaml - Static stick figure
  • stick_figure_waving.yaml - Animated waving
  • curves.yaml - Bezier curves and arcs
  • meeting.yaml - Two figures, one walks to meet the other
  • connection_demo.svg - Architecture diagram using connection elements
  • balloon_friends.svg - Animated stick figures with stretching line

Acknowledgments

This project uses the following open source packages:

  • MCP - Model Context Protocol SDK (MIT)
  • Pydantic - Data validation (MIT)
  • drawsvg - SVG generation (MIT)
  • PyYAML - YAML parsing (MIT)
  • Playwright - Video recording (Apache 2.0)

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

mcp_svg_animator-0.4.1.tar.gz (20.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

mcp_svg_animator-0.4.1-py2.py3-none-any.whl (24.9 kB view details)

Uploaded Python 2Python 3

File details

Details for the file mcp_svg_animator-0.4.1.tar.gz.

File metadata

  • Download URL: mcp_svg_animator-0.4.1.tar.gz
  • Upload date:
  • Size: 20.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for mcp_svg_animator-0.4.1.tar.gz
Algorithm Hash digest
SHA256 e3921fa84c7e37901143435a16843b7c033846fa3e10e3c9363059d5838f8649
MD5 6f1fcd1c706815ea6f8397c0c6667a13
BLAKE2b-256 dd07f264eabf490bf7ff6c9815cf674dbf82a4061f6338ed2c431308822e5398

See more details on using hashes here.

File details

Details for the file mcp_svg_animator-0.4.1-py2.py3-none-any.whl.

File metadata

File hashes

Hashes for mcp_svg_animator-0.4.1-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 de283bbf8e79b57b22a0dbf5bffe27b140c83280a0bdd790e943d3588edef240
MD5 06d67df5ff0c476c805dd7162d8dce90
BLAKE2b-256 2cab9ee83162b2049785f53cc9c37817de933e26658159e44ee73eeafb51c729

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page