WeWork OpenAI Node

? Click to expand the full function screenshot

Enterprise WeChat × OpenAI

WeWork OpenAI (node)

www.connectai-e.com

commercial support

If the open source version cannot meet your company's needs, it is recommended that you try the commercial delivery version currently in internal testing.

• Built-in out-of-the-box Azure Openai: No need to deploy overseas, you can get dozens of times performance improvement
• Admin Panel that controls the overall situation: AI resource management, conversation log query, risk word avoidance and conversation permission management
• Dedicated technical support: Equipped with professional deployment and delivery personnel and one-on-one maintenance services in the future
• Also provides online Saas version/enterprise-level private deployment

View more: https://www.connectai-e.com

?Project features

• Complete enterprise application scenarios,
• One-click deployment, Serverless serverless mode,
• Wizard-based installation, detailed documentation, safe and reliable, low network delay, fast speed

About this project

This project can implement one-click deployment of ChatGPT to Enterprise WeChat, so that ChatGPT and Enterprise WeChat are perfectly integrated. Open Enterprise WeChat on your mobile phone or computer, and you can use the powerful ChatGPT intelligent Q&A. This solution is highly flexible. If you do not have development capabilities, you can use the following method 1 and install it directly with one click. If you are a developer and have secondary development capabilities, you can expand it arbitrarily according to the following method 2.

1. Function integration: Integrate ChatGPT question and answer function into Enterprise WeChat. With the help of Enterprise WeChat permission function, ChatGPT can be shared to enterprise applications. The requirements for this function are simple and only require Enterprise WeChat administrator permission. For the method, please refer to the deployment method below.
2. For more powerful function expansion, this project is an open source project. Partners with development capabilities can Fork to their own warehouses and develop some other functions based on their own business needs, such as combining the open API of Enterprise WeChat.

Deployment method

Cloud function deployment

1. Create an enterprise WeChat application

The first step is to create an application. How to operate: Enterprise WeChat administrator, Enterprise WeChat on the computer-->Avatar-->Manage enterprise-->Application management-->(bottom) Create application-->Application logo + Fill in basic information

The second step is to configure the application. Operation method: Receive message column-->Set up API reception-->Wait for the fourth step before completion.

The third step is to record and prepare the following field information, corpid, agentid, secret, token, aeskey, open-api-key

corpid : Enterprise WeChat on the computer-->Avatar-->Manage Enterprise-->My Enterprise (below) Figure omitted

agentid, secret : Visible after completing the first step.

token, aeskey : In setting API reception (the second step above)

open-api-key needs to be generated in a ChatGPT account (if you don’t have a chatgpt account, you can also ask others to generate one, but this will incur fees)

Application URL API KEY

The fourth step is to deploy to Render with one click (proxy server, free, you can upgrade for a fee)

As shown in the picture, fill in the above field information and click Apply.

You need to wait 3-5 minutes for deployment. After completion, copy the URL of the generated service, as shown below, then copy the URL and append /message . For example, the URL is https://abc.com and splice it into https://abc.com/message. , paste it into the second step page above, and then click Save. When you see a prompt similar to "Save Successfully", it means that the service has been configured successfully!

The last step: Enterprise trusted IP configuration. At the bottom of the application page, configure the trusted IP.

The IP address is as shown below, Render->Connect->Outbound

Configure these three IP addresses into trusted IPs, and now all configuration work is completed.

? Now you can experience it directly. Use Enterprise WeChat on your mobile phone or computer to enter the application you created, and you can chat with ChatGPT.

Server docker deployment

1. Compared with the cloud function deployment solution, this solution requires:

   • a server
   • a domain name

2. Create an enterprise WeChat application (refer to云函数部署1.创建企业微信应用第一、二、三步)

3. Copy the variable file .env.example and fill in your own configuration

   • If the server network and openai api cannot be directly connected, you can configure a custom openai api address, OPENAI_PROXY_URL .
     If no customization is required, please leave it blank.

      OPENAI_PROXY_URL=openai.abc.com

4. Run docker

   Assume the new variable file is named .env.local

    # docker4bill/ww-openai-node:alpine 为构建好的镜像，你也可以利用本仓库的 Dockerfile 构建自己的镜像
   docker run --env-file .env.local -p 6060:6060 -d docker4bill/ww-openai-node:alpine

5. Use caddy or nginx to reverse the above services

   Assume that the domain name is abc.com , take caddy as an example to configure:

    abc.com {
     reverse_proxy localhost:6060
   }

6. Add server IP to enterprise trusted IP

7. Fill in the receiving message server configuration URL

   Assume that the reverse address is abc.com , add /message after it , and fill in the address https://abc.com/message into the URL configured by the message receiving server.