ToolJet/ToolJet

Open-source foundation of ToolJet AI - the enterprise app generation platform for internal tools, dashboards, business applications, workflows and AI agents. Build visually, from a prompt, or from Claude Code, Codex and Cursor over MCP 🚀

Summary

ToolJet is an open-source platform for building internal tools such as admin panels, dashboards, and operational applications on top of existing databases, APIs, and SaaS systems. The repository combines a visual app builder with AI-assisted generation while separating Community Edition capabilities from ToolJet AI Enterprise features.

What it is useful for

ToolJet is useful when engineers need to deliver data-backed business applications without building every screen and interaction from scratch. Its visual builder includes more than 80 responsive components, including tables, charts, forms, lists, and progress bars, plus support for multi-page applications and collaborative editing.

Teams can connect applications to more than 90 data sources, including databases, APIs, cloud storage, and SaaS platforms. ToolJet also includes a built-in no-code database, JavaScript and Python execution inside apps, granular access controls, comments, and mentions. This makes it relevant for internal operations screens, support dashboards, data administration panels, and workflow-oriented applications that need to remain editable by non-specialists after engineering hands them over.

The repository also documents a Model Context Protocol server for building or modifying ToolJet applications through Claude Code, Codex, Grok Build, Cursor, or another MCP-compatible client. Agents work with ToolJet’s component and data contracts, so generated output remains part of the ToolJet app model rather than becoming an unrelated code export.

How engineers can use it

The quickest documented path is to create a ToolJet Cloud account. For self-hosting, the README points engineers to the deployment documentation for Docker, Kubernetes, Helm, AWS, GCP, Azure, and other supported environments.

A local Docker trial is documented directly:

docker run \
  --name tooljet \
  --restart unless-stopped \
  -p 80:80 \
  --platform linux/amd64 \
  -v tooljet_data:/var/lib/postgresql/13/main \
  tooljet/try:ee-lts-latest

After startup, the practical workflow is to connect a data source, define queries, and assemble pages from the visual components. The README also provides examples such as a time tracker, a MongoDB-based CMS, and an AWS S3 browser.

For agent-assisted work, follow the ToolJet MCP setup guide and use a supported client. MCP is explicitly marked as beta, so it should be evaluated with that limitation in mind. The README also places AI app generation, workflows, multi-environment management, and several advanced security features under ToolJet AI Enterprise rather than Community Edition.

Engineers contributing to the codebase should note that the project uses git flow: develop is the base branch, while main and v1.x.x tags represent stable versions. For upgrades, the README recommends the LTS release over latest.