AI Tools
Hermes Agent
Watch an AI agent think, search and report — transparently.
What it does
The useful parts, plainly
Plans before it acts
Hermes breaks your question into steps and shows the plan before running a single search.
Shows its sources
Every claim links back to where it came from, so you can check the work yourself.
Counts the cost
A live token and cost meter runs the whole time — no surprise bills, no hidden spend.
Screenshot / demo GIF
Hermes started as a way to answer one question honestly: what does it actually take for an AI agent to research something well? Most agent demos hide the messy middle. Hermes does the opposite — it shows the plan, the searches, the dead ends, and the bill.
How to try it
- Open the app and type a research question.
- Review the plan it proposes, then let it run.
- Watch the steps and the cost meter as it works.
- Read the report — and click through to the sources to check it.
It’s deliberately unpolished in places. That’s the point: this is what a working agent really looks like under the hood, costs and limits included.
The honest bit — costs & limits
- It costs real money per run — roughly 8–20p depending on depth. The meter shows it live.
- It can still be confidently wrong. Treat it as a fast first draft, not a final answer.
- Long tasks can hit rate limits and pause. It resumes, but it isn't instant.
- There's no login or saved history yet — close the tab and the run is gone.