MCP Catalogs
Homemcp-server screenshot

mcp-server

by bitDive·86·Score 47

MCP server providing trace analysis, request reproduction, and regression management for AI agents.

developer-toolsmonitoringai-llm
20
Forks
0
Open issues
1 mo ago
Last commit
2d ago
Indexed

Overview

The BitDive MCP server exposes monitoring and QA operations to MCP clients like Cursor and Claude Desktop. It implements 23 tools for discovering active modules, fetching traces, reproducing requests, comparing traces before/after code changes, and managing test groups. The server adds important value beyond API proxying including trace readability formatting, comparison logic with normalized Java structures, and reproduction workflow helpers that generate curl/PowerShell commands.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Discover which module, service, or entrypoint is active in a system
you:Compare traces before and after code changes to detect regressions
you:Reproduce captured web requests locally for debugging
you:What does this server provide beyond the BitDive API?
you:How do I configure authentication?

When to choose this

Choose this server if you're working with Java applications and need AI agents to analyze runtime traces, reproduce issues, and compare behavior changes without leaving your development environment.

When NOT to choose this

Avoid this server if you're not using BitDive for monitoring or if you need support for non-Java applications, as it's tightly integrated with the BitDive ecosystem.

Tools this server exposes

12 tools extracted from the README
  • get_heatmap_all_system

    Get a compact heatmap summary for the entire system

  • get_heatmap_for_module

    Get a heatmap summary for a specific module

  • get_heatmap_for_service

    Get a heatmap summary for a specific service

  • get_last_calls

    Get the most recent system calls

  • find_trace_all

    Find traces across the entire system

  • find_trace_for_method

    Find traces for a specific method

  • get_trace_names_batch

    Get trace names in batches

  • get_reproduction_command

    Get reproduction commands for captured requests

  • search_methods_short

    Search for methods with short information

  • compare_traces

    Compare two traces to detect differences

  • create_test_group

    Create a new test group in BitDive

  • get_all_test_scripts

    Get all test scripts from BitDive

Comparable tools

appdynamics-mcpnew-relic-mcpdatadog-mcpstackdriver-mcp

Installation

Install dependencies:

pip install -r requirements.txt

Run in stdio mode:

python server.py

Or in streamable-http mode:

MCP_TRANSPORT=streamable-http MCP_HOST=0.0.0.0 MCP_PORT=8000 python server.py

Configure Claude Desktop:

{
  "mcpServers": {
    "bitdive": {
      "command": "python",
      "args": [
        "/absolute/path/to/server.py"
      ],
      "env": {
        "BITDIVE_MCP_TOKEN": "your-token"
      }
    }
  }
}

FAQ

What does this server provide beyond the BitDive API?
It adds trace readability formatting, comparison logic with normalized Java structures, reproduction workflow helpers, and test management formatted for quick agent use instead of raw responses.
How do I configure authentication?
Set either the BITDIVE_MCP_TOKEN environment variable or pass mcp_token parameter directly to tools. The server requires a valid BitDive MCP token to function.

Compare mcp-server with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.