TAC Insights

TAC Insights is ascript which demonstrate how we can use LLM with Cisco Case API to get data and generate insight from the opened case

Technology stack: Langchain, Python, Cisco Case API, OpenAI model
Status: Alpha

Use Case

TAC insight helps to connect Cisco Case API in meaningful way to identify case, RMA, serial number daata.
It could be used by partner and customer with SNTC or PSS partner

Installation

Detailed instructions on how to install, configure, and get the project running. Call out any dependencies. This should be frequently tested and updated to make sure it works reliably, accounts for updated versions of dependencies, etc.

Sample:

Clone the repo

git clone https://github.com/vivlinux/tac_insight.git

Go to your project folder

cd tac_insight

Set up a Python venv
First make sure that you have Python 3 installed on your machine. We will then be using venv to create an isolated environment with only the necessary packages.

Install virtualenv via pip

pip install virtualenv

Create the venv

python3 -m venv venv

Activate your venv

source venv/bin/activate

Install dependencies

pip install -r requirements.txt

Configuration (optional)

Ensure to copy .env.example file and configure the following:

cp .env.example .env

Set up following details:

Generate Cisco CASE API Client ID and Client Secret
Configure OpenAI Token
Configure OpenAI Model and Version
Configure Webex Token and Email format

Extracting Webex Token

Go to https://developer.webex.com/docs/api/getting-started
Click on "Get Started" and login with your Webex account
Click on "Create a Bot" and follow the instructions
Copy the Bot Token and paste at WEBEX_BOT_TOKEN in the .env file
Copy the email format and paste at WEBEX_BOT_USERNAME in the .env file
Save the .env file

Extract Cisco Case API Client ID and Client Secret

Go to https://apiconsole.cisco.com/
Click on "Create an App" and follow the instructions
Copy the Client ID and paste at CISCO_API_CLIENT_ID in the .env file
Copy the Client Secret and paste at CISCO_API_CLIENT_SECRET in the .env file
Save the .env file

Extract OpenAI Token

Go to https://platform.openai.com/docs/guides/authentication
Click on "Get Started" and follow the instructions
Copy the API Key and paste at OPENAI_API_KEY in the .env file
Provide Model such as gpt-4 or gpt-4o and paste at OPENAI_MODEL in the .env file
Provide Model Version such as 2023-08-01-preview and paste at OPENAI_MODEL_VERSION in the .env file
Save the .env file

Usage

Run the service

make run

How to test the software

Access the bot in Webex Teams
Type help to get the list of commands
Provide query or select query from the list of query