[Showcase] Torra Community Edition – Visual Workflow Builder for AI Agents (Built on LangChain.js)

[HLingSoft]

Hi everyone,

I wanted to share a project I’ve been working on that’s closely tied to LangChain.js — it’s called Torra Community Edition, an open-source visual editor for building AI agents and complex workflows.

Why Torra?
After working with LangChain.js, I realized that designing, chaining, and deploying agent-based workflows could be much more intuitive and accessible — especially for teams who want a no-code/low-code, visual way to manage and audit every step.

What’s unique:

Visual workflow editor (drag & drop, modular, inspired by Langflow)

Full LangChain.js integration: build, connect, and test chains/agents visually

Unified data access layer: GraphQL-style queries, powered by FeathersJS, supports SQLite, MySQL, MongoDB

Powerful hooks: All CRUD/data ops go through FeathersJS hooks for auth, rate-limiting, audit logs, etc.

Multi-language UI, self-hosted, TypeScript fullstack

We built Torra to let teams go from idea → prototype → production much faster, with auditable, reusable workflows.
Currently supports major LLMs (OpenAI, Anthropic, Google, DeepSeek, etc), API connectors, multimodal inputs, built-in runtime, and more.

Would love feedback from the LangChain.js community — especially around advanced agent use cases, data ops, and visualizing chains.
If anyone here has tried to bring more workflow/visual editing to LangChain, I’d love to hear about your experience or needs!

GitHub: https://github.com/HLingSoft/torra-community

Thanks for all your incredible work on LangChain.js — it’s made building these tools so much more fun and flexible.