HN Explorer

Show HN: An n8n alternative where coding agents build the workflows, not humans (velane.sh)

4 points by abhishekraj272 ↗ HN
n8n is built for humans dragging nodes on a canvas. That breaks down at B2B scale (embedding in a product, multi-tenant scalability, etc). n8n does have an MCP server so agents can create workflows too, but it outputs raw JSON. That's fine for n8n's engine, but painful for a coding agent (or a human reviewing its output) to read, write, diff, or debug.

I'm building an alternative where workflows are authored by a coding agent in [a more dev-legible format] instead of JSON blobs, and execute it at scale.

2 comments

[ 2.0 ms ] story [ 15.3 ms ] thread
revanthrj333
Makes sense. I wonder if the next abstraction after "agents generate workflows" is "agents generate intent, and workflows become a compiled artifact."

Feels like manually editing huge JSON graphs is similar to editing compiled code instead of the source.

Interested to see what representation you landed on.

abhishekraj272
"agents generate intent, and workflows become a compiled artifact" -> this was partially done by Composio, more like your AI agent gives command/intent to Composio's agent to get some info, then Composio's creates workflow on the fly and runs it. But it was adopted and they abondoned it (for b2b), maybe it too early.

My understanding was (based on 2yrs of building agents), generate workflows while building the agent, and let it call it.

Also I don't believe in representation of workflow, just let your coding agent create and vaidate it, if it works, it works.