Hi HN,

I'm one of the founders of VoltAgent. Over the past months, we've been building it with a simple focus: AI development lacks real-time observability.

When you work with agents, it's often a black box. You write some code, run it, and just hope it behaves. If something goes wrong, debugging feels painful. Tools like n8n showed us how powerful it is to see a workflow step-by-step, but as soon as we needed customization and deeper control, we hit the limits of no-code tools.

VoltAgent is our attempt to bring that visibility to a real framework. It's an open-source, TypeScript framework for building and orchestrating AI agents with observability built in from the start.

What VoltAgent offers - TypeScript-native building blocks -> tools, memory, multi-agent orchestration, and seamless ai-sdk integration. - Real-time observability -> OpenTelemetry-based tracing and a visual console to inspect how agents think step by step.

Repo: https://github.com/VoltAgent/voltagent Docs: https://voltagent.dev/docs/

We’d love your thoughts, feedback, or questions.

Voltagent’s UI is stunning. Do you have plans to add nocode action types. Sometimes it is easier to start with nocode and switch to coding when necessary.

Really liked the UI and observability, made it way easier to see what was going on.