
Nakkas
Declarative SVG generation where the AI designs the entire config and nakkas renders it. You get a single render_svg tool that accepts a JSON schema covering canvas, gradients, filters, animations (both CSS keyframes and SMIL), and a full element library including shapes, text, groups, and specialized layout groups like radial, grid, and scatter. The AI fills in every field from stroke styles to easing curves while you describe what you want in plain language. Preview returns a PNG, save writes to disk, and everything runs locally with zero external dependencies. Built for AI-native workflows: the schema uses rich describe() annotations so models know exactly what each field does. Good fit when you need programmatic vector graphics without touching Illustrator or writing SVG by hand.
- GitHub stars
- 31
- tools
- 3
- transport
- STDIO
- last updated
- Jul 13, 2026
Tools
3 toolsPublic tool metadata for what this MCP can expose to an agent.
render_svgTakes SVGConfig JSON, returns a PNG preview + artifact id (+ design analysis warnings)
Takes SVGConfig JSON, returns a PNG preview + artifact id (+ design analysis warnings)
No parameter schema in public metadata yet.
previewRe-renders a stored artifact (or raw SVG) to PNG at any width
Re-renders a stored artifact (or raw SVG) to PNG at any width
No parameter schema in public metadata yet.
saveSaves a stored artifact (or raw content) to disk as SVG (text) or PNG (raster)
Saves a stored artifact (or raw content) to disk as SVG (text) or PNG (raster)
No parameter schema in public metadata yet.
Sponsors
Readme
nakkas
MCP server that turns AI into an SVG artist.
One rendering engine. AI decides everything.
Officially listed on the MCP Registry, Glama, LobeHub, and PulseMCP.
Details
- Repository
- github.com/arikusi/nakkas
- Category
- Design & Creative
- Package
nakkas- Transport
- STDIO
- Registry
- MCP registry, active