ToolJet logged 16 release events in 30 days, with 30 GitHub releases over 90 days and a ToolVitals shipping score of 100. That is the useful signal here: ToolJet is not just maintaining an old low-code builder, it is actively reshaping it around AI app generation, workflows, agents, and enterprise internal apps.

The official site now leads with a clear pitch: build full-stack enterprise internal apps in minutes, using natural language to create apps, AI agents, and workflows. The GitHub repository describes ToolJet as the open-source foundation of ToolJet AI, an enterprise app generation platform for internal tools, dashboards, business apps, workflows, and AI agents.

That positioning matches the release stream. Recent LTS and beta releases include upgrades to the AI Coding Assistant, keyword filtering for AI-generated messages, workflow access fixes, workflow build crash fixes, app-specific login and logout flows, and several table-builder improvements.

The release stream says ToolJet is rebuilding the builder layer

The most interesting pattern is not one big launch. It is the mix of AI, workflow, data-source, and component work inside frequent releases.

In v3.20.150-lts, ToolJet added a Tags column type, column pinning for the Table component, and an AI Coding Assistant upgrade. In v3.20.149-lts, it added GUI mode support for MSSQL and keyword filtering for AI-generated messages. In v3.21.24-beta, the changes focused on workflow access across environments and a workflow build crash.

That is a product bet. ToolJet seems to be tightening the boring parts that internal tools need, tables, permissions, data sources, workflow stability, while pushing AI closer to the app-building path. Good internal-app platforms win on these details. Nobody adopts one because the demo is cute if production tables, auth, and database connectors are flaky.

The split between LTS and beta releases also matters. ToolVitals sees 16 release events in 30 days, and the recent list includes both v3.20.x-lts and v3.21.x-beta tags. That suggests a team shipping fixes to a stable lane while moving newer app-builder work through beta.

The GitHub signal is strong, but not the whole story

ToolVitals gives ToolJet a 94 overall score, 100 health score, 100 shipping score, 229.1 hot score, and 37,875 GitHub stars. The official website displayed 37,876 GitHub stars when checked, but ToolVitals metrics are the source of truth for this article, so the post uses 37,875.

The repository is public, and GitHub shows the ToolJet/ToolJet project with 633 tags and more than 12,000 commits on the page I checked. I am not using those as ToolVitals metrics. They are context for why the release cadence looks like a mature open-source project rather than a thin marketing wrapper.

What ToolVitals cannot infer

ToolVitals can see public signals: releases, stars, score movement, website availability, SSL, and activity patterns. It can say ToolJet is shipping often and that recent releases mention AI, workflows, table components, data sources, permissions, and stability fixes.

ToolVitals cannot tell you whether ToolJet’s AI app generation works well for your team. It cannot measure generated app quality, runtime reliability inside your stack, user satisfaction, enterprise renewal rates, or whether the new AI assistant actually improves developer speed.

The safe read is this: ToolJet is alive, maintained, and moving toward AI-assisted enterprise app generation. The unsafe read would be claiming it is the best internal tool platform because it ships often. Shipping is evidence of motion, not proof of product fit.

ToolJet’s 16 release events in 30 days trail some high-velocity neighbors. n8n shows 51 release events in 30 days with 187,130 GitHub stars. LangChain shows 33 release events with 136,160 stars. Gemini CLI is closer on recent activity, with 21 release events and a 229.2 hot score versus ToolJet’s 229.1.

That comparison is useful, but not perfect. n8n is tracked under automation, LangChain under developer tools, and Gemini CLI is a command-line AI tool. ToolJet sits in a different spot: enterprise internal apps, workflows, dashboards, and AI agents in one builder.

React Email is a cleaner contrast for size versus velocity. It has 42 release events in 30 days and 19,131 stars. ToolJet has fewer release events, 16, but more stars, 37,875, and a broader product surface. Bigger surface area usually makes frequent releases harder, not easier.

Recommendation

If your team builds internal tools around databases, workflows, dashboards, and permissions, evaluate ToolJet because the project is actively shipping in the exact layers that decide whether an internal app builder survives production use.

Do not evaluate it only from the AI pitch. Test the table component, workflow builder, data-source GUI modes, auth flows, and deployment model first. The AI layer is the headline, but the release notes say the real work is in making generated internal apps behave like software your company can run.

Sources