Hermes agent makes most of the API based servcies glue into itself very quickly. And you get to work through them just with the prompt. I know on dev.to a lot of people are too much fundamentalist that they prefer code over no-code approach.
However I have learned how to automate better with Hermes than writing from scratch myself. So I found an API which is browseruse which allows you to do browser automation, take screenshot, automate browser for data extraction.
I created a process video for Hermes agent to be integrated with the Browseruse API.
You get 5$ worth of credit. I know lot less than the firecrawl and the browserbase. So that would be one of the blocker. But if you have means to pay for their API you can go ahead and try them out with initial credit.
And if you happen to like their service you can go ahead and build your own service around the same if you can. But paying is worth it if you are into automation. What I have found is that firecrawl and browserbase are enough for most of the tasks.
In case of the browser use the best use case is the pdf export and the screenshot. If you only want these two things then this can be a good API to explore. But apart from that I would say API would do most of the tasks there.
Have you explored any browser automation APIs with hermes agent? What's your experience so far?
Top comments (0)