• REGRESSION
  • ERRORS
  • SUMMARY
  • HUMAN

How to analyze simulation performance results with your AI

Connect Gatling to your AI client to turn raw execution logs into actionable performance insights.

Ask AI about this page

Short answer

How do I analyze Gatling simulation results with AI?

Your AI uses the Gatling MCP to pull specific run data, error rates, and latency metrics. It then compares these numbers against your requirements to spot regressions or failures. You get a clear breakdown of where your simulation went wrong.

Performance outcomes.

Where your data goes.

The AI processes raw Gatling metrics into these specific outcomes.

REGRESSION

Identify performance drops

The AI compares current timing data against previous runs to flag latency increases. You get a direct list of which endpoints slowed down.

ERRORS

Pinpoint failure causes

The AI scans error counts and response codes to find patterns in failed requests. It highlights if a specific service is throwing 500s.

SUMMARY

Get high-level reports

The AI condenses massive execution logs into a brief summary of pass/fail rates and throughput. This is perfect for a quick status check.

HUMAN

Manual code fixes

The AI identifies the bottleneck, but you must still modify your Gatling scripts or application code to fix the underlying issue.

The workflow

What your AI does when the data arrives.

The AI handles the heavy lifting of data retrieval and comparison.

  1. Fetch run history

    The AI looks up your previous simulation attempts to establish a baseline for comparison.

    list_runs

  2. Pull specific metrics

    The AI grabs the exact timing, error, and throughput data for a specific execution.

    get_run

  3. Analyze trends

    The AI compares the retrieved data to find deviations in response times or spikes in error counts.

    get_run

  4. Report findings

    The AI presents the results in a readable format, highlighting the most critical performance issues.

    list_runs

Try it

Copy these to start.

Use these prompts to kick off your analysis.

Starting points

These are starting points. Replace the specific run IDs or names with your actual Gatling data.

Gatling Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_f1mjnAhJFdhCLHf3npA2HLKXXMLgR80q7ymxxmFF/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 — Gatling capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "gatling-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_f1mjnAhJFdhCLHf3npA2HLKXXMLgR80q7ymxxmFF/mcp"
    }
  }
}
Copy into chat04

  • List my last five Gatling runs so I can pick one to analyze.

  • Get the details for run ID 12345 and tell me the total error count.

  • Compare the response times of the most recent run against the one before it.

  • Check the results of the 'LoadTest_Production' run and summarize the throughput.

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

Start here

Connect Gatling once, then ask.

Just link your Gatling instance to the MCP. Your credentials stay encrypted, and you can immediately start asking your AI about your test results.

Connect Gatling to your AI

FAQ

How this task behaves.

  • 01

    Can the AI change my Gatling simulation scripts?

    No. The AI can only read the results of completed runs. You have to manually edit your scripts.

  • 02

    Does the AI delete any of my test data?

    No. The AI only has read access to your Gatling run data through the provided tools.

  • 03

    Can I ask about runs that haven't finished yet?

    No. The tools only work with completed execution data.

  • 04

    How does the AI know which run to look at?

    You tell it. You can use the list of recent runs to identify the ID you want to inspect.

  • 05

    Can the AI trigger a new Gatling simulation?

    No. This MCP is designed for analyzing results, not for initiating new test executions.

  • More questions about Gatling? The Connector page answers them. See everything the Gatling Connector can do

Connect Gatling to Claude, Cursor, ChatGPT & more