Hacker Newsnew | past | comments | ask | show | jobs | submit | Mrakermo's commentslogin

hm, interesting! I like, what I've done is that each step in the process is one agent. One agent get's one task, and only the tool to do it.

So not openclaw, promp injections is a part of the backend based on evalas and scorer meaning that right tool was called, and what each agent can expect.

Instead of having a lot of subagents getting their memory filled with previous runs, prompt injection can be a better way to really narrow down each subagents actual task.


Openclaw is great, but it's still early adopters and often tech savvy people who use it. This is for non tech savvy people in a small companies that are still hesistant to let AI Run their workflow, and n8n and zapier takes too much time to setup and maintain, or the if/then isn't working in their setting.

Openclaw = tech savvy people small team, really knows AI Operator23 = Wants safe agents super easy to setup doing one task and learn about it.


Hey HN,

Most automation tools are painful to set up and break easily.

I built Operator23 to fix that.

You just describe what you want: “Find leads → send emails → wait for replies → follow up → log in CRM”

And it runs automatically across tools like HubSpot, Gmail, Apollo, Notion, etc.

No drag-and-drop. No code. No complex setup.

Key idea:

AI builds the workflow for you It can wait (e.g. for replies) and continue later Works with real, messy workflows (not just simple automations)

Trying to make automation actually usable without spending hours building flows.


Hey HN! I'm Eric, founder of MeetGoran. The problem:Sales managers can only review ~3% of their team's calls. The other 97% of coaching opportunities get lost in transcripts nobody reads.

What we built: An AI that joins sales calls and provides real-time coaching. It tracks frameworks like MEDDIC and BANT live, and prompts reps when they miss key questions.

After calls, it shows managers exactly what each rep needs to improve.

Why it's different from Gong/Chorus: Those are post-call analysis tools. We coach in real-time during the call.

Think of it as a copilot, not a black box recorder.

Tech: Real-time speech recognition, custom NLP models trained on sales frameworks, meeting platform integrations

(Zoom/Teams/Meet).

Current stage: Closed beta with 5 early teams.

Looking for feedback from the HN community.

Website: www:meetgoran.com Promo: https://www.youtube.com/watch?v=mZD56E6lIt4

Happy to answer technical questions!


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: