minware MCP Server

The minware MCP (Model Context Protocol) server lets you connect AI coding tools directly to your minware analytics data. Once configured, your AI assistant can query dashboards, retrieve engineering metrics, and answer questions about your team's performance — without leaving your development workflow.

Prerequisites

  • A minware account with Org Owner permissions
  • One of the supported AI tools listed below

Step 1: Generate an API Key

  1. In minware, go to Settings > API Keys
  2. Click New API Key
  3. Enter a descriptive name for the key (e.g., Claude Desktop or Cursor)
  4. Click Create
  5. Copy the key immediately — it will not be shown again

Note: API keys carry the permissions of the user who creates them. Only Org Owners can create API keys. Treat your API key like a password — do not share it or commit it to version control.

Step 2: Configure Your AI Tool

The minware MCP server is available at:

https://app.minware.com/api/mcp

Select your AI tool below for setup instructions.

Claude Desktop

Edit your Claude Desktop configuration file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

Add the minware entry to the mcpServers object:

{
  "mcpServers": {
    "minware": {
      "url": "https://app.minware.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Restart Claude Desktop for the changes to take effect.

Claude Code

Run the following command to register the minware MCP server:

claude mcp add --transport http minware https://app.minware.com/api/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"

Verify with /mcp in the Claude Code.

Gemini CLI

Add minware to your Gemini CLI settings at ~/.gemini/settings.json:

{
  "mcpServers": {
    "minware": {
      "httpUrl": "https://app.minware.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

For project-level configuration, use .gemini/settings.json in your repository root instead.

Cursor

Open Cursor Settings > MCP, or add minware to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "minware": {
      "url": "https://app.minware.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Restart Cursor after saving.

Codex

Add minware to your Codex configuration at ~/.codex/config.toml. The recommended approach uses an environment variable to keep your API key out of the config file:

[mcp_servers.minware]
url = "https://app.minware.com/api/mcp"
bearer_token_env_var = "MINWARE_API_KEY"

Then export the key in your shell profile (e.g., ~/.zshrc or ~/.bashrc):

export MINWARE_API_KEY=YOUR_API_KEY

Alternatively, you can store the key directly in the config:

[mcp_servers.minware]
url = "https://app.minware.com/api/mcp"
bearer_token = "YOUR_API_KEY"

Example Prompts

Once configured, try prompts like these to explore your engineering data:

Show me the Benchmarks data for the last 30 days. Which metrics need attention?

What was the average cycle time for the backend team last quarter?

Compare sprint performance across teams over the past 3 months.

The assistant will use minware's tools to list teams and dashboards, fetch the relevant data, and provide analysis in natural language.