• CRITICAL
  • TRENDS
  • BACKLOG
  • HUMAN

How to identify technical debt with your AI

Connect Code Climate to your AI client to turn raw code quality data into a prioritized refactoring plan.

Ask AI about this page

Short answer

How can I use AI to find technical debt?

Your AI pulls the latest snapshots and specific code issues from Code Climate to highlight where your codebase is struggling. It then categorizes these issues so you can decide what to fix now and what can wait. You'll get a clear view of your most urgent refactoring needs.

Where the debt lands.

Organized refactoring targets.

The AI sorts findings into these four categories.

CRITICAL

Immediate fixes

The AI flags high-severity issues that threaten stability. You get a list of files that need attention right away.

TRENDS

Growing hotspots

The AI compares recent snapshots to find where debt is accumulating. This helps you spot decaying modules before they break.

BACKLOG

Low priority debt

Minor issues that don't impact current velocity are grouped together. You can review these during slower sprint cycles.

HUMAN

Architectural decisions

The AI flags complex patterns that require a senior dev's judgment. It provides the context, but you make the final call on the fix.

The workflow

What your AI does when the data arrives.

The AI handles the heavy lifting of scanning snapshots and parsing issue lists.

  1. Fetch snapshots

    Your AI looks at the available history for your repository to find the most recent state of the code.

    list_repository_snapshots

  2. Extract issues

    It pulls the specific technical debt markers and error reports from the chosen snapshot.

    list_snapshot_code_issues

  3. Analyze patterns

    The AI looks for recurring problems across different files to find systemic issues.

    list_snapshot_code_issues

  4. Draft plan

    The AI summarizes the findings into a readable list of actionable refactoring tasks.

    list_repository_snapshots

Try it

Copy these to start.

Use these prompts to kick off the process.

Starting points

Swap out the repository names and issue types to match your actual project.

Code Climate Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_dEa3ubWm5DI8grNI5Jrd9891vm9odnIoA9LITGNs/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Code Climate capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "code-climate-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_dEa3ubWm5DI8grNI5Jrd9891vm9odnIoA9LITGNs/mcp"
    }
  }
}
Copy into chat04

  • List the most recent snapshots for the 'api-gateway' repo and tell me which one has the most issues.

  • Show me the code issues from the latest snapshot in the 'frontend-app' repository.

  • Based on the recent snapshots in 'auth-service', what are the top three recurring technical debt issues?

  • Summarize the critical issues found in the last snapshot of the 'data-pipeline' repo.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Code Climate once, then ask.

Just link your account once. Your credentials stay encrypted, and you can immediately start asking your AI about your code quality.

Connect Code Climate to your AI

FAQ

How this task behaves.

  • 01

    Can the AI fix the code issues for me?

    No. The AI identifies and summarizes the issues from Code Climate, but you must perform the actual code changes in your IDE.

  • 02

    Does the AI delete any data in Code Climate?

    No. This MCP only has permission to read snapshots and issue data. It cannot modify or delete anything in your Code Climate account.

  • 03

    How recent is the data the AI sees?

    The AI sees whatever snapshots are currently available in your Code Climate repository via the list tools.

  • 04

    Can I ask about issues in a specific file?

    Yes, once the AI has pulled the issue list for a snapshot, you can ask it to filter or focus on specific files or modules.

  • 05

    Will the AI automatically create tickets in Jira?

    No. This specific connector only reads data from Code Climate. You would need a separate connector to move that data into a task manager.

  • More questions about Code Climate? See everything the Code Climate Connector can do

Connect Code Climate to Claude, Cursor, ChatGPT & more