The FastGPT V4 version has been released. It can organize workflow through Flow visualization to realize complex question and answer scenarios, such as online Google search, operating databases, etc. The function is very powerful. Students who have not used it yet should try it.
Compared with similar products, Feishu is an office tool with a very good experience. Our company also spent a lot of effort to move to Feishu, so let’s access Feishu’s tutorial first.
Original link: https://forum.laf.run/d/1035
Laf part
The first is the function code of Laf. Just like before connecting to Enterprise WeChat, I have prepared a template for everyone. You only need to fill in the environment variables to use it.
Here laf.dev is used to demonstrate laf.run. The operation is the same. It is just that during the recent special period, laf.run is the domestic server calling FastGPT. The Singapore server is very slow.
Log in to laf.dev and find the function template: FastGPT access Feishu
After clicking on the template, click "Use Template" in the upper right corner to add this template to your application.
Then we clicked "Settings" and "Environment Variables" and found that there are four empty environment variables here. Fill them in and we're done.
Feishu backstage
We go directly to Feishu Open Platform and click Create Application.
Here "Create enterprise self-built application" fill in the information and then create it.
Here we want to add robot capabilities to this application.
Then here is "Permission Management", configure the permissions according to the picture below, there are five permissions in total.
Event subscription here, you need to add an "accept message"
Then come to event subscription and configure the "request address"
The request address here is the Laf cloud function address you just created, fill it in and save it.
Go back to the voucher and basic information again, get AppID
andApp Secret
Fill in the environment variables of Laf FEISHU_APPID
andFEISHU_SECRET
After completing this, the application can be released. Let's create a version in "Version Management and Release" and fill in the version number and introduction.
The available range here needs to be customized by yourself and which colleagues can use it.
Then save it and let colleagues with authority review it.
FastGPT section
Come to FastGPT and find the application you want to access. Like me, get AppId
and ApiKey
fill in the Laf environment variables respectively.FASTGPT_APPID
FASTGPT_KEY
Remember to click "Update" after filling in
use
After passing the review, we found the application on the workbench and tested it in a private chat.
ok. We have it all done now. We can chat privately and join group chats. There is no IP whitelist and it is very smooth to use.