bzm

The BlazeMeter MCP Server creates a bridge between AI tools and BlazeMeter's cloud-based performance testing platform, enabling agents and chatbots to orchestrate load testing workflows. It supports multiple deployment methods (Docker, uv/uvx for Python-based installs, manual binary usage, and more) and exposes a configurable MCP client setup that can be loaded into editors like Cursor or VS Code. After you set up the MCP with your BlazeMeter API credentials, you can invoke BlazeMeter actions such as creating tests, launching runs, monitoring results, and generating reports directly from natural language prompts. The included tooling allows you to either run the server in a container (Docker), install from source (uv/uvx with Python 3.11+), or use pre-built binaries for a quick start. The configuration examples show how to wire environment variables (e.g., BLAZEMETER_API_KEY in various forms) to point to your API key JSON file and to specify working directories for test files. Once configured, you can add the MCP to your editor clients to manage BlazeMeter workflows with AI-assisted prompts, retrieve status updates, and perform operations via the MCP UI or integrated command links.

Prerequisites:

• Docker (recommended for quick setup) or Python 3.11+ with uv/uvx for source installations
• BlazeMeter API credentials (api-keys.json) saved in the same folder as your MCP binary or accessible to your environment
• Optional: uv/uvx installed if you plan to install from source code

Installation steps (Docker-based):

1. Ensure Docker is installed and running on your machine.
2. Obtain your BlazeMeter API keys JSON (api-keys.json) and place it in the same folder as your MCP configuration or binary.
3. Run the MCP container with the appropriate environment and mount points. Example:

docker run --rm -i \
  --mount type=bind,source=/path/to/your/test/files,target=/home/bzm-mcp/working_directory/ \
  -e API_KEY_ID=your_api_key_id \
  -e API_KEY_SECRET=your_api_key_secret \
  -e SOURCE_WORKING_DIRECTORY=/path/to/your/test/files \
  ghcr.io/blazemeter/bzm-mcp:latest

Manual installation from source (uv/uvx and Python 3.11+):

1. Install uv (https://docs.astral.sh/uv/) and Python 3.11+ on your system.
2. Install the MCP from the repository or a released version, following the uv/uvx instructions in the docs.
3. Configure the MCP client with your api-key.json path, as shown in the README examples.

Manual binary installation:

1. Download the binary for your OS from the Releases page.
2. Run the binary once to print a JSON config and links to add the MCP to Cursor/VS Code.
3. Alternatively, manually create an mcpServers entry using the provided JSON example, pointing the command to the binary path and setting the appropriate env variables (e.g., BLAZEMETER_API_KEY to your api-key.json path).

Tips and notes:

• Always ensure your BlazeMeter API credentials are secured. Use a path to api-key.json referenced in env so the MCP can load API keys correctly.
• When using Docker, remember to mount a working directory and propagate necessary environment variables (API_KEY_ID, API_KEY_SECRET, SOURCE_WORKING_DIRECTORY).
• If you’re on macOS and encounter a security prompt for the MCP binary, follow the system prompt to allow the app in Security & Privacy settings.
• The uvx-based installation runs the package in a temporary environment; you can pin to specific releases or branches as needed.
• For Windows users, ensure paths in JSON configs use escaped backslashes where required (e.g., C:\path\to\dir).
• You can add MCP configurations to your editor clients (Cursor or VS Code) via the provided add-to-clipboard/config links or by manually editing the mcpServers JSON.
• If using Docker with a custom CA, you may need to add SSL_CERT_FILE settings or mount certificate bundles as described in the Custom CA Certificates section of the README.