taigas-texture
MCP server for procedural material generation and inspection via a deterministic Texture DSL, exposing tools to list, validate, render, and multiscale-inspect materials.
README
<p align="center"> <img src="./assets/banner.png" width="100%" alt="TAIGA Texture"> </p>
<p align="center"><code>deterministic procedural materials · browser workbench · CLI · MCP</code></p>
TAIGA Texture is a compact procedural-material laboratory with a deterministic Texture DSL that can be authored by people or driven by an LLM.
TTX is the technical shorthand taken from Taiga TeXture. It names the file format and toolchain; the language itself is called Texture DSL.
Edit materials in the browser, inspect PBR channels and a cube preview, then automate the same engine through the CLI or MCP.
[ MATERIAL GALLERY ]
<table> <tr> <td align="center"><img src="./assets/gallery/crackle-ceramic.png" width="220"><br><sub>CRACKLE CERAMIC</sub></td> <td align="center"><img src="./assets/gallery/saddle-leather.png" width="220"><br><sub>SADDLE LEATHER</sub></td> <td align="center"><img src="./assets/gallery/quarter-sawn-oak.png" width="220"><br><sub>QUARTER-SAWN OAK</sub></td> </tr> <tr> <td align="center"><img src="./assets/gallery/granite.png" width="220"><br><sub>GRANITE</sub></td> <td align="center"><img src="./assets/gallery/waxed-canvas.png" width="220"><br><sub>WAXED CANVAS</sub></td> <td align="center"><img src="./assets/gallery/rusted-steel.png" width="220"><br><sub>RUSTED STEEL</sub></td> </tr> </table>
Features
- Texture DSL — pure, forward-only fields with deterministic seeds and compile-time cost limits.
- Material workbench — live source editing, PBR channel views, tiling checks, parameter sliders, and an interactive studio-lit cube.
- PBR ZIP export — download the current browser material as Base Color, Height, OpenGL Normal, Roughness, Metallic, metadata, and its
.ttxsource. - Procedural library — ceramic, leather, stitched leather, oak, granite, canvas, rusted steel, marble, and multiscale parched earth.
- LLM tool surface — structured validation diagnostics, deterministic analysis, PNG rendering, multiscale inspection, CLI, and stdio MCP.
- No runtime packages — the browser workbench is one HTML file; the automation tools use Node.js built-ins.
Run locally
Clone the repository and open ttx-bench.html in a WebGL 2 capable browser.
git clone https://github.com/kawaiiTaiga/taigas-texture.git
cd taigas-texture
On Windows:
Start-Process .\ttx-bench.html
Optional verification:
npm test
Export to Blender, Unity, Unreal, or Godot
In the browser workbench, choose the preview resolution and press PBR export → download .zip. The archive contains five standard PNG maps, material.json, a short integration note, and the authoritative .ttx source.
Use Base Color as sRGB. Load Height, Normal, Roughness, and Metallic as Non-Color/Linear data. The normal map follows OpenGL (+Y); invert its green channel for a DirectX normal workflow.
CLI
# Discover bundled and library materials
node ttx-cli.mjs examples --json
# Validate one example
node ttx-cli.mjs validate --example saddle --json
# Render PBR views
node ttx-cli.mjs render --example ceramic --out renders/ceramic --resolution 256 --json
# Inspect the same point across scales
node ttx-cli.mjs render-scales materials/parched-earth.ttx \
--out renders/parched-earth --zooms 1,4,16,64 --center 0.5,0.5 --json
MCP
Start the dependency-free stdio server:
node ttx-mcp-server.mjs
Example MCP configuration:
{
"mcpServers": {
"taigas-texture": {
"command": "node",
"args": ["<ABSOLUTE_PATH>/ttx-mcp-server.mjs"]
}
}
}
Exposed tools:
list_material_examplesget_material_examplevalidate_materialrender_materialrender_material_multiscale
Material example
seed = 17
tile = 1
var glaze_h = 0.53
field cells = worley(u, v, freq=15, jitter=1, out=edge, seed=3)
field crack = 1 - smoothstep(0.004, 0.052, cells)
field h = clamp(0.52 - 0.11*crack, 0, 1)
out height = h
out albedo = hsv(glaze_h, 0.42, 0.72*(1 - 0.48*crack))
out rough = clamp(0.20 + 0.44*crack, 0, 1)
out metal = 0
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.