- HEALTH
- COVERAGE
- COMPLEXITY
- HUMAN
Short answer
How can I check my code quality with an AI?
Your agent uses DeepSource to pull real-time metrics like test coverage and repository health. You can ask for specific reports or high-level health summaries to find where your code needs work.
Where every metric lands.
The outcomes of your queries.
Your agent processes the data and organizes it into these specific areas.
HEALTH
Repository health checks
The AI pulls status reports to show you if your codebase is stable or accumulating technical debt.
COVERAGE
Test coverage insights
Get exact percentages on how much of your code is actually covered by tests.
COMPLEXITY
Code complexity analysis
Identify overly complicated functions or files that are becoming hard to maintain.
HUMAN
Manual code review
The AI flags the issues, but you still need to jump into the editor to actually fix the logic.
The workflow
What your AI does when the data arrives.
The agent acts as the bridge between your chat interface and your DeepSource dashboard.
Fetch health data
The agent requests a full overview of the repository's current state.
get_report_cardAnalyze coverage
It looks up specific test coverage numbers to see where gaps exist.
get_test_coverageCheck complexity
The agent pulls metrics regarding how difficult the code is to read or maintain.
get_repository_metricsSummarize findings
It combines all the retrieved data into a plain English summary for you.
get_report_card
Try it
Copy these to start.
Use these prompts to get moving immediately.
Starting points
Swap out the repository names for your actual project names.
DeepSource Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_gRojrYBqBJYk2MB0kmqFd1Bq7M9JWeWmipQYTeBA/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — DeepSource capabilities are ready to use.
{
"mcpServers": {
"deepsource-mcp": {
"url": "https://edge.vinkius.com/vk_preview_gRojrYBqBJYk2MB0kmqFd1Bq7M9JWeWmipQYTeBA/mcp"
}
}
}Give me a report card for the backend-api repository.
What is the current test coverage for the auth-service repo?
Show me the repository metrics for the frontend-app to check for complexity.
Which repositories in my DeepSource account have low test coverage?
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect DeepSource once, then ask.
Just link your account through the Vinkius dashboard. Your credentials stay encrypted, and you can start querying your metrics immediately.
Connect DeepSource to your AIFAQ
Common questions
- 01
Can the AI fix the code issues it finds?
No. The AI can identify and report on issues, but you must manually apply the fixes in your codebase.
- 02
Does the AI have permission to delete my repositories?
No. The tools provided only allow the agent to read metrics and reports. It cannot perform destructive actions like deleting code or repos.
- 03
Can I see coverage for a specific file?
Yes, you can ask the agent to check coverage for specific parts of your repository.
- 04
How often is the data updated?
The AI pulls the most recent data available in DeepSource at the moment you ask the question.
- 05
Can the AI create new tests for me?
No. The agent can only retrieve existing coverage data; it cannot write or execute new test files.
More questions about DeepSource? The Connector page answers them. See everything the DeepSource Connector can do
Connect DeepSource to Claude, Cursor, ChatGPT & more
