1
- # ConvoStack
1
+ < h1 style = " margin-top : 0 px " > ConvoStack 💬⭐</ h1 >
2
2
3
3
<img width =" 1439 " alt =" Screenshot 2023年06月02日 at 2 47 17 AM " src =" https://github.com/ConvoStack/convostack/assets/8688852/40fdbf63-4f53-4143-8628-39c934eaaed4 " >
4
4
5
- ## What is ConvoStack?
5
+ ## Key Features
6
6
7
- ConvoStack is a plugandplay embeddable AI chatbot widget and backend deployment framework for your website. ** It is
7
+ ConvoStack is a plug- and- play embeddable AI chatbot widget and backend deployment framework for your website. ** It is
8
8
completely free and open source and currently running on our [ docs] ( https://docs.convostack.ai/ ) website!**
9
9
10
10
The core technologies are:
@@ -21,28 +21,27 @@ the [docs](https://docs.convostack.ai/).
21
21
22
22
To see a live demo of ConvoStack, check out our free [ playground] ( https://playground.convostack.ai ) !
23
23
24
- ## Learn the Basics
25
-
26
- Learn about ConvoStack and what sets it apart from other AI chatbot software:
27
-
28
- - [ Why ConvoStack?] ( https://docs.convostack.ai/the-basics )
29
-
30
24
## Getting Started
31
25
32
- Get your AI chatbot up and running in minutes with our [ quickstart guide] ( https://docs.convostack.ai/getting-started/ ) :
26
+ Get your AI chatbot up and running in minutes with our [ Quickstart repo and guide] ( https://github.com/ConvoStack/quickstart ) :
33
27
34
28
In the following example, we are connecting a Langchain OpenAI [ LLM] ( https://js.langchain.com/docs/modules/models/llms/ ) to the chatbot playground.
35
29
36
30
``` typescript
31
+ import * as dotenv from " dotenv" ;
32
+ // Configures the OpenAI API key
33
+ dotenv .config ();
34
+
37
35
import { playground } from " convostack/playground" ;
36
+ import { IAgentContext , IAgentResponse } from " convostack/agent" ;
38
37
import { OpenAI } from " langchain/llms/openai" ;
39
38
40
39
playground ({
41
- reply(context : IAgentContext ): Promise <IAgentResponse > {
40
+ async reply(context : IAgentContext ): Promise <IAgentResponse > {
42
41
// `humanMessage` is the content of each message the user sends via the chatbot playground.
43
42
let humanMessage = context .getHumanMessage ().content ;
44
43
// `agent` is the OpenAI agent we want to use to respond to each `humanMessage`
45
- const agent = new OpenAI ();
44
+ const agent = new OpenAI ({ modelName: " gpt-3.5-turbo " } );
46
45
// `call` is a simple string-in, string-out method for interacting with the OpenAI agent.
47
46
const resp = await model .call (humanMessage );
48
47
// `resp` is the generated agent's response to the user's `humanMessage`
0 commit comments