API Key for Self Hosted Firecrawl
Hi,
I have deployed the self host version of Firecrawl in my system to link it with Dify. In Dify, it requires API Key to integrate. In self host version, we don't have any API Keys or is there a way to generate API Keys. Please help me to solve this issue.
Thanks in advance
14 Replies
On the self host you can bypass the need for an API key by disabiling authentication by setting
USE_DB_AUTHENTICATION=false
in your .env
. Then i believe it should just work out of the box in DifyIn Dify, we need to enter API Key and it is a mandatory field
I have attached the screenshot for configuring Firecrawl in Dify

Try setting the base url to your local one and on the API key field just writen anything and see if it works. Your localhost should be bypassing the need for the API key so anything you input there should technically work.
Hey @Adobe.Flash I'm stuck on the same thing as @Saran , I'm certain that the api is up and running correctly cause I get sucessful responses with postman. When I click on "save" on Dify, it stays loading for a long time, and after a while, I see on chrome's dev console that the dify call gets a "status code of 524". I tried debuggin with gpt4turbo and it says:
I'm running the local firecrawl with Docker
I don't have any error logs from the docker container
so, gpt4turbo concluded that the problem is the local instance that is taking to long to send back a reply to Dify, which is weird, can I can send jobs through postman that run flawlessly
Also I pulled the latest release of firecrawl, and rebuild the docker image
Hi @grooveseeker , interesting ccing @rafaelmiller here too... I am assuming you have tried what I said above, correct?
Just to recap:
-
USE_DB_AUTHENTICATION=false
in your .env
- Setting Dify API Key to any text you'd like (fc-abcdefghijk)yes, I tried that. I also tried using the same firecrawl api key (just to make sure I followed the same character structure), even tho it shouldn't matter since I have the USE_DB_AUTHENTICATION=false in my .env
Are you using the python sdk? If yes, you need to specify an api key - https://github.com/mendableai/firecrawl/blob/main/apps/python-sdk/firecrawl/firecrawl.py#L33.
GitHub
firecrawl/apps/python-sdk/firecrawl/firecrawl.py at main · mendable...
🔥 Turn entire websites into LLM-ready markdown or structured data. Scrape, crawl and extract with a single API. - mendableai/firecrawl
I wasn't using the python sdk, I was trying to connect Dify to my local instance of firecrawl. Would it be possible to use the sdk with a local instance?
hello guys, did any of you find a solution to this?
Hey @grooveseeker we contacted Dify about this.
Still haven't heard back
We will follow up with them
The same problem exists with Flowise, except Flowise doesn't allow the base url to be adjusted.
@BrianJM Let me reach out to them and see if we can get the base url to be adjusted there!
Thanks @Adobe.Flash. I opened a thread on the Flowise Discord asking if it was possible (maybe I overlooked it).
hii, have you got dify working with selfhosted firecrawl yet?
Hi, I return this parameter when I use the local connection Dify.The local project is already deployed, and I guess it might be an API problem. But the local deployment projects should not need api?Thanks
