/
San Francisco
/
We added Steel as a cloud browser provider in Hermes, the open source AI agent from Nous Research. New SteelProvider, a steel_scrape tool, and setup wizard support. Set STEEL_API_KEY, pick steel as your browser provider, and Hermes routes all browser sessions through Steel with proxy and CAPTCHA solving built in.
You can try it today from the branch, or wait for the merge into main. We couldn't wait. Our team is already running it in their deployed Hermes agents, and we took it for a spin to show what it looks like.
Setup
Run hermes setup. The parts that matter:
Pick your model.
Under browser automation, select Steel.
Paste your Steel API key. Get one free at app.steel.dev.
What happened
We ran Hermes and gave it one task:
Find me the best hotel near Grand Central New York and show me Steel's session URL so we can observe what the agent is doing.
Agent started working. It went to google.com/travel, searched for hotels, then navigated to Booking.com on its own. Nobody told it to go there. It decided that's where the data was.
This part was slow. LLMs are slow between actions. The agent was navigating, clicking through hotels, ruminating. The web is hard. For doing anything meaningful on it, you need a proper browser.
While this was happening, the live session was visible in Steel's dashboard.
Watching it work
You don't need to watch every click. But when something goes wrong, when the agent misclicks or gets lost in some modal, you need to see what it was looking at. That's the difference between debugging and guessing.
The integration pipes Steel's live viewer URL back into Hermes responses, so you get the link right in the agent's output. And if the agent hits bot detection, it knows to start stealth session or fall back to steel_scrape instead of retrying the same approach.
The agent built its own skill
After finishing the hotel search, Hermes created a skill called "hotel search" on its own. Nobody asked it to. It had lived through the whole navigation sequence and decided to package that up for next time.
We've seen the same pattern with Claude Code and Codex. Manual run, then script, then automation. The difference here is Hermes skipped being asked and just did the second step itself.
The scrape tool
The PR also includes steel_scrape, a dedicated tool for server-side content extraction. No browser session needed. Hit Steel's /v1/scrape API and get back pure markdown.
Markdown that an agent can actually consume. Give it a list of URLs, have it ingest them, build a knowledge base. The scrape tool works separately from browser sessions, so use whichever fits.
More than one session
This was one agent in one session. Steel runs browsers in the cloud. Let loose a fleet of agents doing parallel sessions at the same time.
With Steel you can bring your own credentials. You can keep sessions alive. Import local browser profiles with steel profile import and the agent inherits whatever accounts you're already logged into. Or create a dedicated profile with limited access.
Any agent, any model
Hermes is open source, in our case we are running an open model from Z.ai. Steel doesn't care what's behind the decisions. GLM 5.1 worked. GPT would work. Llama would work. If your agent can make tool calls, it can use Steel.
The browser is the browser.
Try it
Try it from the branch today, or wait for the merge into main.
All Systems Operational