Skip to main content
Table of Contents

MCP Server

The LinearB MCP (Model Context Protocol) server lets you query your LinearB engineering data from supported AI tools and IDE assistants using natural language. Connect your preferred MCP-compatible c…

Table of Contents

The LinearB MCP (Model Context Protocol) server lets you query your LinearB engineering data from supported AI tools and IDE assistants using natural language. Connect your preferred MCP-compatible client to explore delivery, quality, throughput, DevEx, and AI Analytics insights without navigating multiple dashboards.

TL;DR

  • Connect supported MCP clients—including Claude, Codex, Cursor, VS Code, and others—to LinearB.
  • Use OAuth for secure, user-based authentication. API keys remain available for clients that do not support OAuth.
  • Query LinearB metrics, repositories, contributors, releases, incidents, teams, pull requests, project management data, and AI Analytics.
  • Use natural-language questions or ready-made prompts to generate reports and investigate engineering trends.
  • Always verify important results in LinearB before using them for critical decisions or official reporting.
Ready to connect? Follow the step-by-step instructions in Connect to LinearB MCP Server.

Overview

The LinearB MCP server is a secure gateway between your MCP-compatible client and your LinearB organization. After connecting, your client can call LinearB tools to retrieve structured engineering data and use that data to answer your questions.

You can use the MCP server to:

  • Explore delivery, quality, throughput, DevEx, and DORA metrics.
  • Investigate pull requests, repositories, contributors, releases, incidents, services, teams, and branches.
  • Query project management data such as issues, epics, features, and initiatives.
  • Explore AI adoption, usage, and delivery-impact data available through AI Analytics.
  • Generate custom reports using natural-language questions.

Availability and Requirements

  • The LinearB MCP server is generally available to Company Admins.
  • You need an MCP-compatible client.
  • OAuth is recommended and does not require creating or storing a LinearB API key.
  • An organization-scoped LinearB API key can be used when your client does not support OAuth or requires static credentials.
  • The MCP endpoint is https://mcp.linearb.io/mcp.

How the MCP Server Works

1. Connect and authenticate

Add the LinearB MCP endpoint to your client. Depending on the client, authentication begins through a Connect prompt, an automatically opened browser, or a command-line login step.

OAuth redirects you to LinearB to sign in and approve access. If OAuth is unavailable in your client, you can configure an API key instead.

2. Ask a natural-language question

For example:

  • Show me cycle time trends for the frontend repository over the last six months.
  • List incidents resolved by Team Alpha this week.
  • Generate a merge rate report for all backend teams for the previous quarter.
  • Compare AI-assisted and non-AI-assisted pull request cycle time.

3. Your client calls LinearB tools

Your AI client translates the request into one or more MCP tool calls. LinearB returns the relevant structured data, and the client uses it to prepare a response.

4. Review and verify the result

Because the client uses generative AI to interpret your question and summarize the returned data, responses may occasionally be incomplete or inaccurate. Verify important figures in LinearB before publishing a report or making a critical decision.

Example Use Cases

Engineering leaders
  • Benchmark delivery, quality, and throughput across teams.
  • Identify patterns shared by high-performing teams.
  • Find repositories, workflows, or teams with persistent bottlenecks.
  • Explore how AI adoption relates to delivery and quality outcomes.
Team leads and managers
  • Break down coding, pickup, review, and deployment time.
  • Find urgent or long-running pull requests.
  • Review delivery trends for a specific team, service, or repository.
  • Prepare data-informed retrospectives and coaching conversations.
Executives and cross-functional partners
  • Generate tailored reports without navigating dashboards.
  • Answer ad hoc questions about cycle time, merge rate, and release outcomes.
  • Filter results by team, contributor, repository, or timeframe.
  • Summarize engineering trends for business reviews.

Available Data

The MCP server provides tools for data domains including:

  • Engineering metrics and benchmarks
  • Repositories and contributors
  • Pull requests and branches
  • Teams, groups, and users
  • Releases, incidents, and services
  • Issues and project management entities
  • AI development tools and AI Analytics

Available tools may evolve as LinearB adds new data and capabilities.

Prompt Library

LinearB provides ready-made prompts for common reports, including:

  • Merge rate report — Explore code delivery velocity and bottlenecks.
  • Lead time report — Measure how long work takes from start to completion.
  • Urgent PR report — Identify high-priority pull requests that need attention.
  • Release report — Retrieve pull requests associated with specific releases.

Download the LinearB MCP Server Prompt Library for ready-to-use examples.

Accuracy and Responsible Use

  • Be specific about the timeframe, team, repository, and metric in your question.
  • Review the filters and assumptions used in the response.
  • Verify important results in the LinearB Dashboard before using them for official reporting or business decisions.
  • Do not include credentials, API keys, or other secrets in prompts.

Troubleshooting

The client cannot find the LinearB server
  • Confirm that the MCP endpoint is https://mcp.linearb.io/mcp.
  • Confirm that the server is enabled in your client.
  • Restart the client after changing its MCP configuration.
  • Review the client-specific instructions in Connect to LinearB MCP Server.
Authentication repeatedly fails
  • Confirm that you are signing in to the correct LinearB organization.
  • Confirm that you have the required Company Admin access.
  • If you use an API key, verify that it is valid and belongs to the correct organization.
  • If the key was rotated or revoked, update your client configuration.
The response is incomplete or unexpected
  • Ask one question at a time and specify a clear timeframe.
  • Include the relevant team, repository, or service name.
  • Compare the result with the corresponding data in LinearB.
  • If the issue persists, save the prompt and response and contact LinearB Support.

How did we do?

LinearB MCP Server Connectivity Architecture

MCP Server – Kubernetes Secrets

Contact