Kiln
Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more.
- 5.2k stars
- 395 forks
- 56 open issues
- Language: Python
- License: NOASSERTION
- updated
Install
local · uvx (python)claude mcp add kiln -- uvx kiln{
"mcpServers": {
"kiln": {
"command": "uvx",
"args": ["kiln"]
}
}
}{
"mcpServers": {
"kiln": {
"command": "uvx",
"args": ["kiln"]
}
}
}- Run the command above in your terminal (Claude Code), or paste the JSON config into claude_desktop_config.json (Claude Desktop).
- Replace any <placeholder> values with your API keys or paths.
- Restart Claude. The MCP server and its tools appear automatically.
Package name inferred from the repository name. Verify it exists on PyPI, or clone https://github.com/Kiln-AI/Kiln and follow its README.
Editor's note
Kiln is a desktop workbench and open-source Python library covering the full AI development cycle, including an Eval Builder that auto-generates judge models and synthetic evaluation datasets, an Auto-Optimize feature that tests prompt variants, model selection, and agent configurations to find the best-performing setup, drag-and-drop RAG creation from PDFs, images, video, and audio, zero-code fine-tuning across 60-plus models on Fireworks, Together, and Vertex with serverless deployment, and multi-agent composition via subagents each running in isolated context windows. It exposes an MCP server so Claude Code and other MCP-compatible clients can interact with Kiln tasks and tools directly. The desktop app runs on Mac, Windows, and Linux with one-click installation, stores project data in a Git-native format so non-technical teammates like product managers and domain experts can rate outputs and contribute datasets without writing code, and supports fully offline use via Ollama. The MIT-licensed Python library lets agents built in the app ship straight to production.
MCP Servers overview
What people ask about Kiln
What is Kiln-AI/Kiln?
Kiln-AI/Kiln is mcp servers for the Claude AI ecosystem. Build, Evaluate, and Optimize AI Systems. Includes evals, RAG, agents, fine-tuning, synthetic data generation, dataset management, MCP, and more. It has 5.2k GitHub stars and its last recorded update is dated 2026-10-23.
How do I install Kiln?
You can install Kiln by cloning the repository (https://github.com/Kiln-AI/Kiln) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is Kiln-AI/Kiln safe to use?
Our security agent has analyzed Kiln-AI/Kiln and assigned a Trust Score of 87/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains Kiln-AI/Kiln?
Kiln-AI/Kiln is maintained by Kiln-AI. The last recorded GitHub activity is dated 2026-10-23, with 56 open issues.
Are there alternatives to Kiln?
Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.