OpenGlaze

OpenGlaze is a free, open-source, self-hosted ceramic glaze calculator, recipe manager, and UMF analyzer for potters, ceramic artists, and studios.

TL;DR: OpenGlaze — open-source ceramic glaze calculator (UMF, CTE, recipe optimizer) you can self-host. Best for potters, ceramic artists, studios, and educators.

---

Screenshots

Screenshots of the current app are pending — see docs/screenshots/README.md for the capture list and instructions.

---

Public Discovery

OpenGlaze is open-source glaze chemistry software for potters, ceramic artists, studios, educators, and production ceramicists. It combines a ceramic glaze calculator, UMF analysis, CTE estimation, recipe management, self-hosting, and optional AI glaze consulting.

AI discovery: llms.txt provides a compact project summary for AI assistants and search crawlers.

Best-fit searches: ceramic glaze calculator, UMF calculator, glaze chemistry software, open source pottery app, self-hosted glaze recipe manager, CTE glaze calculator, Glazy alternative, ceramic recipe optimizer.

Overview

OpenGlaze is a free, open-source ceramic glaze management system for potters, ceramic artists, and studios. It combines a UMF calculator, CTE analysis engine, computational recipe optimizer, and AI-powered glaze consulting — all in one self-hosted application.

30-second example: diagnose a crazing glaze

1. Enter a glaze recipe such as feldspar/silica/whiting/kaolin.
2. OpenGlaze calculates UMF, oxide roles, SiO₂:Al₂O₃ ratio, and estimated CTE.
3. If the glaze fit looks risky, the optimizer suggests material adjustments for the next test tile.
4. You still fire a real test tile — but you waste less clay, material, and kiln space getting there.

Popular starting points: Ceramic glaze calculator · UMF calculator · CTE calculator · Glazy companion · Self-hosted glaze software

Built by potters, for potters. Your glaze recipes stay on your infrastructure. No subscriptions. No feature gates. MIT licensed.

Why OpenGlaze?

• 🏠 Own your data — Self-host on your own server. Recipes never leave your studio.
• 🔓 Truly open — MIT licensed, no proprietary lock-in, full source code available
• 🧮 Computational chemistry — UMF calculator, CTE prediction, and recipe optimizer using stoichiometric analysis — not guesswork
• 🧠 AI-powered — Kama assistant understands glaze chemistry and answers technical questions
• 🎨 Studio-ready — Multi-user collaboration, role-based access, shared glaze libraries
• ☕ Free forever — No paywalls, no subscriptions, no feature gates. Voluntary support only.

What is OpenGlaze?

OpenGlaze is a ceramic glaze calculator and recipe management tool designed for anyone working with glazes:

• Hobby potters who want to understand their glazes at the molecular level
• Studio ceramicists managing dozens of recipes and test firings
• Educators teaching glaze chemistry with transparent, reproducible tools
• Production potters optimizing recipes for consistency and cost

At its core, OpenGlaze automates the chemistry that ceramicists have historically done by hand or with spreadsheets: calculating Unity Molecular Formula (UMF), estimating thermal expansion coefficient (CTE), and suggesting material adjustments to achieve target properties.

How It Compares

Tool	What It Does	How OpenGlaze Fits
Glazy	Recipe database & community	Export from Glazy → analyze & optimize in OpenGlaze
DigitalFire	Ceramic chemistry education & reference	Deep educational content; OpenGlaze is the practical calculator
INSIGHT	Desktop glaze calculation software	Similar chemistry engine; OpenGlaze is web-based, open-source, and self-hosted
HyperGlaze	Recipe database	Import recipes into OpenGlaze for UMF analysis

OpenGlaze is not a replacement for these tools — it complements them. Many users maintain their recipe libraries in Glazy and use OpenGlaze for computational analysis and optimization.

Quick Start

Docker (Recommended, 2 minutes)

The default Docker path is a single-user, self-hosted SQLite install with persistent Docker volumes. It does not require PostgreSQL or Ory Kratos.

# Clone the repository
git clone https://github.com/KyaniteLabs/openglaze.git
cd openglaze

# Copy environment file and set a real SECRET_KEY before public use
cp .env.example .env

# Start OpenGlaze
docker compose up -d

# Access at http://localhost:8768
curl http://localhost:8768/health

Manual Installation

# Install dependencies
pip install -r requirements.txt

# Run (the SQLite database is created and seeded with the community glazes on first start)
python server.py

Open http://localhost:8768 in your browser. (Docker default; manual install defaults to 8767.)

CLI, MCP, and Agent Skill

OpenGlaze also ships agent-friendly local surfaces:

python -m openglaze_cli brief
python -m openglaze_cli umf --recipe "Custer Feldspar 45, Silica 25, Whiting 18, EPK 12" --cone 10
python -m openglaze_cli batch --recipe "Custer Feldspar 45, Silica 25, Whiting 18, EPK 12" --size 500
python -m openglaze_mcp

• CLI: python -m openglaze_cli exposes project brief, UMF, batch scaling, and substitutions.
• MCP: python -m openglaze_mcp starts a stdio MCP server with glaze recipe tools for agent hosts.
• Skill: skills/openglaze/SKILL.md guides compatible agents on when to use the CLI or MCP server and how to keep glaze-safety claims bounded.

Example MCP config:

{
  "mcpServers": {
    "openglaze": {
      "command": "python",
      "args": ["-m", "openglaze_mcp"]
    }
  }
}

Features

🎨 Glaze Management

• Store unlimited glazes with chemistry, recipes, and visual references
• Track family relationships and base types
• UMF calculator with automatic oxide analysis and SiO₂:Al₂O₃ ratio
• Recipe scaling and batch calculator with cost estimation
• Food safety and cone range annotations (cone 06 through 10)

🔬 Chemistry Engine

• UMF calculator — Unity Molecular Formula from any batch recipe
• CTE analysis — Thermal expansion coefficient prediction for crazing/shivering risk
• Glaze compatibility analysis and material substitution suggestions
• Recipe optimizer — Computational suggestions to hit target CTE, surface (matte/glossy), or durability
• Batch reporting with cost estimation
• Oxide analysis with automatic role classification (flux, stabilizer, glass former)

🧠 AI Assistant (Kama)

• Context-aware glaze consulting
• Streaming responses for real-time help
• Local LLM support (Ollama) or cloud (Claude)
• Chemistry-aware prompt injection
• Experiment suggestion engine

🧪 Experiment Pipeline

• 6-stage workflow: Ideation → Prediction → Application → Firing → Analysis → Documentation
• Photo documentation at each stage
• Structured firing log integration
• Result comparison and archiving
• Reproducibility tracking

👥 Studio Collaboration

• Studio groups and invite-code joining
• Lab assignment tracking
• Shared experiment views
• Simple local-token identity for lightweight collaboration
• Role-based authorization and comment threads are roadmap items

🎮 Gamification

• Points and streak tracking
• Achievement badges
• Activity leaderboards
• Experiment milestones
• Community challenges

More Features

• 🧮 Recipe Optimizer — Computational glaze recipe optimizer suggesting exact material adjustments to hit target CTE, surface, or durability without physical testing
• 📸 Photo Documentation — Gallery view across multiple firings
• 🔥 Firing Logs — Atmosphere, cone, and schedule tracking
• 🧮 Layering Tracker — Document and predict base/top combinations
• 💾 Import/Export — JSON/CSV export in the legacy dashboard; Glazy/INSIGHT import is a roadmap integration
• 📊 Progress views — Track experiments and prediction activity where auth is enabled
• 📱 PWA — Install as an app on mobile/desktop
• ⌨️ Command Palette — Quick navigation with ⌘K
• 🔐 Auth — simple local auth for studio features; Ory/Kratos support is experimental

Learn by search intent

These pages are built to answer the exact questions potters and AI answer engines ask:

Search intent	OpenGlaze page
Ceramic glaze calculator	https://kyanitelabs.github.io/openglaze/ceramic-glaze-calculator.html
UMF calculator for ceramics	https://kyanitelabs.github.io/openglaze/umf-calculator.html
Glaze recipe calculator	https://kyanitelabs.github.io/openglaze/glaze-recipe-calculator.html
Glaze CTE calculator	https://kyanitelabs.github.io/openglaze/cte-glaze-calculator.html
Glazy alternative / companion	https://kyanitelabs.github.io/openglaze/glazy-alternative.html
DigitalFire companion	https://kyanitelabs.github.io/openglaze/digitalfire-companion.html
Open-source pottery software	https://kyanitelabs.github.io/openglaze/open-source-pottery-software.html
Self-hosted glaze software	https://kyanitelabs.github.io/openglaze/self-hosted-glaze-software.html

AI crawlers and answer engines can use docs/llms.txt, docs/llms-full.txt, and docs/ai.txt for canonical project facts.

Tech Stack

Component	Technology	License
Backend	Flask 3.x (Python)	MIT
Frontend	Vanilla JS SPA	MIT
Database	SQLite	Public Domain
Auth	Simple local tokens; Ory hooks experimental	Apache 2.0
AI	Ollama (local) / Anthropic Claude (cloud)	—
Import/Export	JSON/CSV export in legacy dashboard; broader Glazy/INSIGHT import is roadmap	MIT
Chemistry	Custom UMF Engine	MIT
Container	Docker + Compose	—

Architecture

┌─────────────────────────────────────────────────────────────┐
│                        Client Layer                          │
│  ┌──────────────┐  ┌──────────────┐  ┌──────────────┐      │
│  │   Browser    │  │   PWA App    │  │   API Clients│      │
│  └──────────────┘  └──────────────┘  └──────────────┘      │
└────────────────────┬────────────────────────────────────────┘
                     │
┌────────────────────▼────────────────────────────────────────┐
│                     Flask Application                        │
│  ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐       │
│  │  Glazes  │ │Chemistry │ │    AI    │ │ Templates│       │
│  └──────────┘ └──────────┘ └──────────┘ └──────────┘       │
│  ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐       │
│  │ Studios  │ │   Auth   │ │Analytics │ │  Uploads │       │
│  └──────────┘ └──────────┘ └──────────┘ └──────────┘       │
└────────────────────┬────────────────────────────────────────┘
                     │
┌────────────────────▼────────────────────────────────────────┐
│                      Data Layer                              │
│  ┌──────────┐  ┌──────────┐  ┌──────────────────────────┐  │
│  │SQLite    │  │Ory       │  │Ceramics Foundation Data  │  │
│  │          │  │Kratos    │  │(Materials, Recipes, etc.)│  │
│  └──────────┘  └──────────┘  └──────────────────────────┘  │
└─────────────────────────────────────────────────────────────┘

Documentation

Document	Description
Installation	Docker and manual setup guides
Configuration	Environment variables and settings
API Reference	REST API endpoints and examples
Architecture	System design and data flow
User Guide	End-user feature walkthrough
Development	Contributing and local dev setup
Self-Hosting	Production deployment guide
Support	Voluntary support for the project
Changelog	Version history and release notes

Self-Hosting

OpenGlaze is designed for self-hosting. All components are open source.

Minimum Requirements

• 1 CPU core
• 1 GB RAM
• 10 GB storage
• Docker & Docker Compose

Deployment Options

Platform	Cost	Difficulty	Best For
VPS (Hetzner, DO, Linode)	$5-20/mo	Medium	Full control
Render / Railway / Fly.io	$7-25/mo	Easy	Managed platforms
Raspberry Pi / Bare Metal	$0	Hard	Offline/local use

See docs/self-hosting.md for detailed deployment instructions for each platform.

Support the Project

OpenGlaze is free and open source. If it saves you materials, time, or a failed kiln load, consider starring the repo or contributing a recipe.

No pressure — the tool is yours either way.

Customizing Glaze Collections

Glaze templates live in core/templates/ and templates/. To add your own:

1. Create a YAML file following the community-glazes.yaml format
2. Add glaze entries with name, cone, atmosphere, base_type, recipe, etc.
3. Import it via the API, or add it to the first-start seeding logic in server.py (seed_database, which currently loads core/templates/community-glazes.yaml)

Importing via the API needs no code changes.

Data

The ceramics-foundation/ directory contains canonical ceramic reference data:

• 30+ materials with oxide analyses and aliases
• 19 oxides with molecular weights, roles, and safety ratings
• Firing schedules for cone 06 through 10
• UMF target ranges and surface prediction thresholds
• Layering rules and material substitutions
• Studio recipes in YAML format

All data is versioned, sourced, and cited. See ceramics-foundation/ for the full dataset.

Templates

Pre-built glaze collections:

• Cone 10 Reduction (Community) — 28 traditional reduction glazes
• Cone 6 Oxidation (Community) — 12 electric kiln glazes
• Cone 6 Reduction (Community) — 4 gas kiln reduction glazes

See templates/ and ceramics-foundation/recipes/.

Testing

# Run the full test suite
pytest tests/ -v

# Run with coverage
pytest tests/ --cov=. --cov-report=html

The test suite covers:

• Kama AI context injection and streaming
• Chemistry context retrieval and UMF calculation
• Recipe optimizer (target CTE, surface, alkali, running risk)
• Flask route imports and response formats
• System prompt generation and database schema

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

Quick start:

# Fork and clone
git clone https://github.com/YOUR_USERNAME/openglaze.git
cd openglaze

# Install dependencies
pip install -r requirements.txt

# Run tests
pytest tests/

# Make changes and submit a PR

Security

See SECURITY.md for our security policy and vulnerability reporting process.

Support

Questions, bug reports, and feature requests:

• GitHub Discussions — Community help
• GitHub Issues — Bug reports and feature requests

License

• Code: MIT License — Free to use, modify, distribute
• Templates: CC-BY-4.0 — Free with attribution
• Documentation: CC-BY-4.0

Acknowledgments

• The ceramics community for sharing knowledge and recipes
• Contributors who have helped build OpenGlaze
• Ceramic Arts Network for glaze chemistry references
• Digitalfire for oxide data and UMF methodology

---

---

Part of KyaniteLabs

More from KyaniteLabs. Related projects:

• Innerscape — personal-growth OS: journaling & reflection
• Elixis — local-first AI pattern-synthesis engine for ideas
• kinocut — guardrailed video-editing MCP server for AI agents

→ More at kyanitelabs.tech

What is OpenGlaze?

OpenGlaze is a free, open-source, self-hosted ceramic glaze calculator and recipe manager: UMF analysis, CTE estimation, a computational recipe optimizer, and an optional AI glaze assistant (Kama).

Product	OpenGlaze
Category	open-source ceramic glaze calculator and recipe manager
Best for	potters, ceramic artists, studios, and glaze-chemistry educators
Not	a hosted recipe-sharing community or a replacement for real test tiles
Source	GitHub · Forgejo (private, maintainers only)
Keywords	ceramic glaze calculator, UMF calculator, CTE glaze calculator, self-hosted glaze software, Glazy companion

Who it's for

• Primary: potters, ceramic artists, studios, and glaze-chemistry educators
• Use when you need to calculate UMF and CTE, manage glaze recipes, and get computational recipe adjustments on your own infrastructure
• Skip if you need a hosted recipe-sharing community or a replacement for real test tiles

FAQ

What is OpenGlaze?

OpenGlaze is a free, open-source, self-hosted ceramic glaze calculator and recipe manager: UMF analysis, CTE estimation, a computational recipe optimizer, and an optional AI glaze assistant (Kama).

Who should use OpenGlaze?

Potters, ceramic artists, studios, and glaze-chemistry educators.

How is OpenGlaze different?

Unlike recipe databases, OpenGlaze runs the glaze chemistry (UMF, CTE, optimizer) on your own server; it complements Glazy, DigitalFire, and INSIGHT rather than replacing them.

Is OpenGlaze production software?

Treat the README status and release tags as source of truth for maturity. Validate against your own requirements before production use.

Status

• Maintained as of 2026 on the default branch
• Prefer release tags when pinning dependencies
• Report issues on GitHub

Agent surface

• Coding agents: read this README first, then repo docs/AGENTS.md if present
• Prefer machine-readable briefs (llms.txt) when the repo ships one
• MCP or skill entrypoints are documented in-repo when applicable

Contributing

Issues and PRs welcome on GitHub. Keep public docs free of secrets and machine-local paths.

License

See LICENSE in this repository (or package metadata if license is package-only).