Skip to content
MCP server

Agile Analytics MCP

Read-only Azure DevOps sprint, cycle time, WIP and forecast metrics for AI assistants

Runs locallyStatus and history
Install
Pricing
Paid
Platforms
CLI, self-hosted
Runs
Locally
Version
6.79.1
Listing updated
today

Similar MCP servers

About Agile Analytics MCP

Agile Analytics MCP is a paid MCP server for cloud and infrastructure. It runs locally on your machine and is started by your AI client.

This connector is part of the Agile Analytics extension for Azure DevOps and only answers if that extension is installed in your organisation, the organisation is on a trial or has a valid licence, and a workflow mapping is saved for the project. It runs the extension's own metric code locally, so the numbers match what the extension displays.

Authentication uses a personal access token with three read scopes (Work Items, Project and Team, Extension Data), or an Azure CLI sign-in for organisations connected to Microsoft Entra. The README says it sends nothing to Baytek, uses a fixed list of read requests, and never reads titles, assignees, tags, descriptions or comments, only fields such as state, type, dates, estimates, board column and the blocked flag.

You name the organisation, project and team in the prompt. Each answer also reports whether sub-teams are included and how many other teams' items were left out, and it returns workflow-not-mapped rather than guessing a mapping.

Install

Run in a terminal

claude mcp add agile-analytics-mcp --env AZURE_DEVOPS_PAT='<AZURE_DEVOPS_PAT>' -- npx -y @baytekdev/agile-analytics-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "agile-analytics-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@baytekdev/agile-analytics-mcp"
      ],
      "env": {
        "AZURE_DEVOPS_PAT": "<AZURE_DEVOPS_PAT>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "agile-analytics-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@baytekdev/agile-analytics-mcp"
      ],
      "env": {
        "AZURE_DEVOPS_PAT": "<AZURE_DEVOPS_PAT>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "agile-analytics-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@baytekdev/agile-analytics-mcp"
      ],
      "env": {
        "AZURE_DEVOPS_PAT": "<AZURE_DEVOPS_PAT>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "agile-analytics-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@baytekdev/agile-analytics-mcp"
      ],
      "env": {
        "AZURE_DEVOPS_PAT": "<AZURE_DEVOPS_PAT>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "agile-analytics-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@baytekdev/agile-analytics-mcp"
      ],
      "env": {
        "AZURE_DEVOPS_PAT": "<AZURE_DEVOPS_PAT>"
      }
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Configuration

  • AZURE_DEVOPS_PATAzure DevOps personal access token with Work Items: Read, Project and Team: Read and Extension Data: Read. Optional when the Azure CLI is signed in.

Tools this server provides (6)

Taken from the project documentation; the running server may differ.

sprint_summary
Delivered items and points, cycle time, carryover and commitment.
cycle_time
Cycle time over the last N finished sprints.
flow_efficiency
Flow efficiency over the last N finished sprints.
wip
Items in progress in the running sprint.
aging
Open items with stage and days in flow.
forecast
Monte Carlo forecast by items or by date.

What it does

Sprint summary
Delivered items and points, carryover and commitment reliability.
Cycle time and flow
Metrics for the last 3, 6 or 12 finished sprints.
Work in progress
Lists items in progress in the running sprint.
Item aging
Shows stage and days in flow per open item.
Forecasting
Monte Carlo forecast of how many items or when a backlog finishes.
Read-only access
Limited to a fixed set of read requests.

Good for

  • Prepare a sprint review with delivered points and carryover
  • Check which open items are aging in the current sprint
  • Compare cycle time across recent sprints
  • Forecast how much the team can finish in 30 days

Pricing

Paid

Part of an Agile Analytics subscription; trial or licence required

Agile Analytics MCP FAQ

What do I need first?

Node.js 22 or later, the Agile Analytics extension in your Azure DevOps organisation and a saved workflow mapping for the project.

Which token scopes are required?

Work Items: Read, Project and Team: Read, and Extension Data: Read.

Can I avoid creating a token?

Yes, if your organisation uses Microsoft Entra you can sign in with the Azure CLI and leave the token unset.

Does it read work item titles or comments?

No. The README says it never reads titles, assignees, tags, descriptions or comments.

Is Agile Analytics MCP free?

No. Agile Analytics MCP is a paid product.

How do I add Agile Analytics MCP to Claude Code?

Run this command in your terminal: claude mcp add agile-analytics-mcp --env AZURE_DEVOPS_PAT='<AZURE_DEVOPS_PAT>' -- npx -y @baytekdev/agile-analytics-mcp

What are the best alternatives to Agile Analytics MCP?

The closest alternatives to Agile Analytics MCP are Ha MCP, Agent-Native Analytics, Google Ads + Meta Ads + SEO/GEO MCP, mcp and dashboard-icons.

Where this comes from

  • Description: written from the project's README.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers