Using AI to Enhance Existing Enterprise Tools Like Jira

A developer recently demonstrated a practical use case for AI coding agents: enhancing existing enterprise applications rather than building new ones from scratch. Using Claude's Chrome extension, they created a Jira sidebar that displays cross-project dependency graphs—functionality that Jira typically requires multiple clicks and page loads to access.

How It Works

The developer used Claude's Chrome extension, which can inject panels directly into web pages. With just 4 prompts, they created a working sidebar that integrates with their actual Jira instance. The extension reads data from the DOM and augments the existing interface.

Key Points from the Discussion

• Most developers spend their workday inside enterprise applications they didn't choose (Jira, Salesforce, Workday, ServiceNow) that are entrenched for years
• Browser extensions provide a way to enhance these tools without replacing them
• One commenter noted that LLMs excel at creating small programs that integrate with existing APIs
• There's concern that companies like Salesforce and LinkedIn might try to lock down access to their interfaces

Technical Details

The discussion includes practical insights about browser extension development:

• Browser extensions can be prototyped quickly—one developer created a working extension in 20 minutes with AI assistance
• The MDN tutorial for basic browser extensions (injecting JavaScript into pages matching URL patterns) takes less than 10 minutes
• Most extensions are open source, making it easy to modify existing code
• Extensions work by reading what's already in the DOM and augmenting it

This approach represents a shift from the common focus on building complete new applications with AI. Instead, it demonstrates how AI can be used to improve the daily experience of working with existing enterprise tools that aren't going away anytime soon.