Skip to main content

How does Jira Copilot understand your Jira workspace?

What is the knowledge base?

A knowledge base is a self-serve online library of information about a product, service, department, or topic. The knowledge base used in Jarvis Copilot for Jira comes from tickets’ data in a Jira workspace that installed our Jira Copilot. This helps our artificial intelligence (AI) greatly enhance the performance and capabilities in several ways: deeper understanding of specific domains, prompt response to information requests, decision support and data analysis.

How does the knowledge base Work?

The knowledge base leverages advanced language processing technology to assist users. Specifically, we use a powerful AI model called GPT (Generative Pre-trained Transformer) to analyze tickets in a Jira workspace and generate relevant suggestions.

How do we build?

Here's an overview of how the knowledge base is built:

  • Data Extraction: We automatically pull the necessary information from Jira workspace to fuel the knowledge base, such as:
    • All created tickets
    • Any custom templates or writing guidelines user’s organization uses
  • Prompt Engineering: We carefully craft "prompts" that combine the data from Jira workspace with the specific context needed to produce helpful suggestions, such as for user ticket summaries and descriptions.
  • Suggestion Models: We feed these prompts into the GPT language model. This allows the model to understand the user team's unique context and provide personalized writing guidance.
  • Continuous Improvement: As your Jira environment evolves, we regularly update the knowledge base, including refining the prompts for the GPT model, to keep the suggestions fresh and relevant.

How to ensure security?

We take the security and privacy of user information extremely seriously. When users use the Jira Copilot knowledge base, they can be assured that:

  • We only access the necessary data from their organization's Jira workspace, with stringent access controls in place.
  • All data is encrypted in transit and at rest, protecting it from unauthorized access.
  • We fully comply with the user company's security policies and relevant industry regulations.
  • We do not store or retain any data pulled from user Jira workspace in our databases. The information is used solely to provide suggestions.