AVGE Engine

AVGE Engine

Vector illustration tools for AI agents — create scenes, characters, and graphics with shapes, armatures, materials, and rendering.

Category
Visit Server

README

AVGE Engine — AI-Native Vector Graphics Engine

Version: 0.5.0 | Tool set: m0b-v1

Vector illustration tools for AI agents — create editable vector art with shapes, armatures, materials, and rendering. Output is clean SVG you can open in Illustrator, Figma, or any vector editor. Runs as an MCP server.

View examples — fridge scene, bedroom, landscape, iPhone mockup, cat manga, and more with previews.

Quick Start

python -m avge_engine api       # FastAPI HTTP server (:8000)
python -m avge_engine mcp-sse   # MCP SSE server (:8001)
python -m avge_engine mcp       # MCP stdio server

Preview renders at http://localhost:8000/preview/<document_id>.png

Features

  • Normalized coordinates — 0.0–1.0 canvas, no pixel math
  • Procedural geometry — radial spread, offset outlines, guide lines, building chains, speech bubbles, bursts, armature skeletons, foreshortening
  • Armature skeletons — node-edge graphs with tapered segments, junction separation, filleted V-gaps, curved Catmull-Rom chains, and unary union merging
  • Boolean operations — union, intersect, subtract, xor with Ramer-Douglas-Peucker simplification
  • HSL shading — auto highlight + shadow from light direction
  • Primitives — rects (including tapered/trapezoid), ellipses, lines, polylines, arcs, polygons, stars
  • Text + images — SVG text with font family/style/anchor, embedded images via <image>
  • Gradient backgrounds — linear and radial gradient definitions
  • Style system — fill, stroke, stroke-width, opacity, blend modes, dash patterns
  • Batch operations — execute multiple tools in one call, edit multiple regions with per-region transforms
  • Palette generation — HSL harmony presets (complementary, triadic, analogous, etc.)
  • Cross-document copy — copy elements between documents with offset
  • SVG arc support — A/a commands in import_svg_path (elliptical arcs with sweep/large-arc flags)
  • Relative positioning — place elements relative to a parent region's bounding box
  • PNG rasterization — via rsvg-convert (librsvg) with Unicode/emoji font support
  • Per-document tracking — checkpoint/restore history, tool usage stats

42+ MCP Tools

Category Tools
Document create_document, list_documents, set_background, get_document_stats
Create create_region, create_primitive, create_curve, create_text, insert_image, import_svg_path
Edit edit_region (point-level nudge), edit_regions (batch transforms), delete_region, copy_element
Transform transform_objects, duplicate, boolean_operation, mirror_region
Style restyle, add_shading, generate_palette, define_gradient, apply_line_hierarchy, compare_style_consistency
Groups group_regions, ungroup_regions
Procedural generate_shape (14+ patterns: armature, segmented_chain, radial_spread, speech_bubble, create_burst, foreshorten, etc.)
View render_preview, describe_scene, checkpoint_diff, render_diff
History checkpoint, restore, get_history, batch

Coordinates

All coordinates are normalized 0.0–1.0 where (0, 0) = top-left and (1, 1) = bottom-right.

Relative Positioning

Create tools accept relative_to — pass a region ID and coordinates become 0–1 fractions of that region's bounding box:

{"tool": "create_primitive", "shape": {"type":"ellipse","cx":0.5,"cy":0.5,"rx":0.1}, "relative_to": "belt_panel"}

Architecture

avge_engine/
├── controllers/      # MCP tool definitions (region, scene_ops, style, procedural, ...)
├── scene/            # Scene graph (graph.py, models.py)
├── geometry/         # Curve fitting, procedural patterns, types
├── effects/          # Style dataclass, HSL color transforms
├── renderer/         # SVG serializer + PNG rasterization
├── services/         # Engine service, document persistence
├── api.py            # FastAPI HTTP server
└── server.py         # MCP server setup (FastMCP)
  • Python 3.12 with FastMCP (MCP SDK)
  • Normalized coordinates, resolved at render time
  • Catmull-Rom → cubic Bézier curve fitting (deterministic, closed-form)
  • SVG output with text, images, gradient defs
  • PNG preview via rsvg-convert with fontconfig/Unicode support
  • JSON file persistence per document

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured