---
component: ROOT
version: "2026.3"
slug: ROOT/integrations/index
canonical_url: "https://docs.develocity.ai/2026.3/integrations/agentic-ai/mcp-servers/"
title: "MCP Server"
description: "Develocity exposes your build data to AI agents through tools via the Model Context Protocol, for analyzing individual builds and detecting aggregate trends."
keywords:
  - "MCP server"
  - "Build Scan"
  - "failure analytics"
status: current
---

<!-- llms-index: https://docs.develocity.ai/llms.txt -->

# MCP Server

<a id="preamble"></a>

Develocity exposes your build data to AI agents through MCP tools, usable in any MCP-enabled AI agent via the standard [Model Context Protocol (MCP)](https://modelcontextprotocol.io/). The tools come in two sets: the Develocity tools for analyzing individual builds, and the Develocity Analytics tools for querying aggregate data across your organization. AI agents authenticate with a Develocity access key, or by signing in through an OAuth authorization server that Develocity is configured to trust.

<a id="develocity-tools"></a>

## Develocity Tools

The Develocity tools give an AI agent direct access to build context: exception details, stack traces, test outcomes, cache performance, and more. Developers and build engineers use them to investigate build failures, identify test patterns, and compare builds without leaving their AI agent. They bundle skills: guided workflows that activate automatically to help the AI agent make effective use of available data.

Follow the [installation instructions](https://docs.develocity.ai/2026.3/integrations/agentic-ai/mcp-servers/installation/#develocity-mcp-server) to enable them.

> [!NOTE]
> The Develocity tools do not support Bazel builds.

**Investigate Build Issues** — Access exception details, stack traces, environment context, and the build’s console log to understand recent build failures.

**Advanced Failure Analytics** — Use Develocity’s failure grouping (automatic clustering of related build failures) to identify root causes and common failure patterns.

**Investigate Test Issues** — Access the exception, stack trace, and captured output of an individual test execution, including attempts that failed before a retry passed.

**Analyze Test Patterns** — Identify flaky tests, discover failure trends, and analyze test performance across your test suites.

**Monitor Build Performance** — Query execution times, resource usage, network activity, and build caching effectiveness.

**Compare Builds** — Compare two builds to pinpoint input differences and diagnose cache misses.

**Explore Build Data** — Filter across projects, users, build outcomes, time, and custom tags for Gradle, Maven, sbt, npm, Python, and Docker builds.

<a id="develocity-analytics-tools"></a>

## Develocity Analytics Tools

The Develocity Analytics tools give an AI agent access to aggregate build data across your organization. Use them for organization-wide queries: dependency risk assessment, build performance trends, and CI stability analysis. They also bundle skills that activate automatically to guide the AI agent’s queries.

These tools require Develocity 2025.4 or higher with [Athena Data Export](https://docs.develocity.ai/reporting-aws/2.1/) enabled, or [Develocity Reporting Kit](https://docs.develocity.ai/reporting-kit/2.4/) 2.1 or higher. Follow the [installation instructions](https://docs.develocity.ai/2026.3/integrations/agentic-ai/mcp-servers/installation/#develocity-analytics-tools) to enable them.

**Analyze Aggregate Build Data** — Query and analyze large batches of build data across your organization.

**Identify Dependency Risks** — Detect projects using outdated or vulnerable dependency versions.

**Monitor Build Performance Trends** — Analyze configuration times, task durations, and build performance patterns over time.

**Assess CI Stability** — Identify failure patterns, prioritize fixes, and understand the impact of failing builds across users and projects.

<a id="where-the-tools-are-served"></a>

## Where the Tools Are Served

Depending on your deployment, your AI agent can reach these tools differently:

*   On a Develocity instance, a single MCP server at the `/mcp` endpoint serves the Develocity tools. When the instance is configured with an analytics source — either Develocity Reporting via [Athena Data Export](https://docs.develocity.ai/reporting-aws/2.1/) or a connected [Develocity Reporting Kit](https://docs.develocity.ai/reporting-kit/2.4/) — it also serves the Develocity Analytics tools from the same `/mcp` endpoint, subject to the user’s permissions.
    
*   The embedded Analytics MCP endpoint on the Reporting Kit instance (`/drv-mcp`) is deprecated in Develocity 2026.3 and will be removed in a future release. During 2026.3 it forwards requests to the consolidated Develocity MCP Server, so existing clients keep working. Migrate AI agents to the Develocity `/mcp` endpoint instead. See [Changes](https://docs.develocity.ai/2026.3/operations/upgrade/changes/) for migration guidance.
    

You can use either tool set on its own; you don’t need both to get started. Some data, such as test results and failure groups, is only available from the Develocity tools.

<a id="get-started"></a>

## Get Started

*   [Installation Manual](https://docs.develocity.ai/2026.3/integrations/agentic-ai/mcp-servers/installation/): Enable the MCP server on your Develocity instance
    
*   [Authentication](https://docs.develocity.ai/2026.3/integrations/agentic-ai/mcp-servers/authentication/): Configure the credentials the MCP server accepts
    
*   [AI Agents](https://docs.develocity.ai/2026.3/integrations/agentic-ai/mcp-servers/ai-agents/): Configure your AI agent to connect to the MCP server