#
Configuring ai tools
We use three main AI helpers at DataWars:
continue.dev
for VSCodesgpt
for command line helpjupyter-ai
Private API Keys to setup AI are located in the following document
Danger
Don't share the API Keys with ANYBODY. This is highly sensitive information.
#
Setting up continue.dev
If you have any doubts, go to the Continue docs
Here's a quick setup guide:
For the most part, you can use the gpt-4o-mini
, as it's a fast, effective and cheap model. For more advanced tasks you can use gpt-4o
, but it's more expensive. Use it with discretion.
The config file for continue.dev will look something like:
{
"tabAutocompleteModel": {
"apiKey": "API Key",
"title": "Codestral",
"model": "codestral-latest",
"provider": "mistral",
"apiBase": "https://api.mistral.ai/v1"
},
"models": [
{
"model": "gpt-4o-mini",
"title": "GPT-4o Mini",
"systemMessage": "You are an expert software developer. You give helpful and concise responses.",
"apiKey": "API KEY",
"provider": "openai"
},
{
"model": "gpt-4o",
"title": "GPT-4o",
"systemMessage": "You are an expert software developer. You give helpful and concise responses.",
"apiKey": "API KEY",
"provider": "openai"
}
],
"contextProviders": [
// The rest of the confs...
],
"slashCommands": [
// other confs...
],
"data": []
}
#
SGPT
SGPT docs: https://github.com/TheR1D/shell_gpt
SGPT is very simple to install, just pip install shell-gpt
.
Here's a quick start guide: