# Koog — Kotlin/JVM Agent Framework with MCP Tools

> A JVM agent framework for predictable, fault-tolerant agents, with MCP tool integration; verified 4187★, pushed 2026-05-14.

## Install

Copy the content below into your project:

## Quick Use

```kotlin
// Gradle (Kotlin DSL) from README
dependencies {
  implementation("ai.koog:koog-agents:0.7.3")
}
```

## Intro

A JVM agent framework for predictable, fault-tolerant agents, with MCP tool integration; verified 4187★, pushed 2026-05-14.

**Best for:** JVM teams building production-grade agents who want typed, structured integrations

**Works with:** Kotlin/Java builds (Gradle/Maven); integrates MCP tools per README overview

**Setup time:** 12-25 minutes

### Key facts (verified)

- GitHub: 4187 stars · 403 forks · pushed 2026-05-14.
- License: Apache-2.0 · owner avatar + repo URL verified via GitHub API.
- README-backed entrypoint: `ai.koog:koog-agents:0.7.3`.

## Main

- Start as a library, not a template: add Koog via Maven/Gradle, then grow your agent surface incrementally.

- Use MCP as the boundary for tools: keep tool definitions structured so your agent logic stays testable.

- Prefer predictable control flow: treat retries/timeouts as first-class, especially when adding external tools.

- Document the setup in your repo: dependency + minimal example beats a long, vague architecture note for onboarding.

### README (excerpt)

**Koog**

[![Kotlin Beta](https://kotl.in/badges/beta.svg)](https://kotlinlang.org/docs/components-stability.html)
[![Maven Central](https://img.shields.io/maven-central/v/ai.koog/koog-agents)](https://search.maven.org/artifact/ai.koog/koog-agents)
[![JetBrains incubator project](https://jb.gg/badges/incubator.svg)](https://github.com/JetBrains#jetbrains-on-github)
[![Kotlin](https://img.shields.io/badge/kotlin-2.2-blue.svg?logo=kotlin)](http://kotlinlang.org)
[![CI status](https://img.shields.io/github/checks-status/JetBrains/koog/main)](https://github.com/JetBrains/koog/actions?query=branch%3Amain)
[![GitHub license](https://img.shields.io/github/license/JetBrains/koog)](LICENSE.txt)

Build status:

[![Checks](https://github.com/JetBrains/koog/actions/workflows/checks.yml/badge.svg?branch=develop)](https://github.com/JetBrains/koog/actions/workflows/checks.yml?query=branch%3Adevelop)
[![Heavy Tests](https://github.com/JetBrains/koog/actions/workflows/heavy-tests.yml/badge.svg?branch=develop)](https://github.com/JetBrains/koog/actions/workflows/heavy-tests.yml?query=branch%3Adevelop)
[![Ollama Tests](https://github.com/JetBrains/koog/actions/workflows/ollama-tests.yml/badge.svg?branch=develop)](https://github.com/JetBrains/koog/actions/workflows/ollama-tests.yml?query=branch%3Adevelop)

Useful links:

* [Documentation](https://docs.koog.ai/)
* [API reference](https://api.koog.ai/)
* [Slack channel](https://docs.koog.ai/koog-slack-channel/)
* [Issue tracker](https://youtrack.jetbrains.com/issues/KG)

## Overview

Koog is a Kotlin-based framework designed to build and run AI agents entirely in idiomatic Kotlin and Java API. It lets you create agents that can interact with tools, handle complex workflows, and communicate with users.

### Key features

Key features of Koog include:

- **Multiplatform development**: Deploy agents across JVM, JS, WasmJS, Android, and iOS targets using Kotlin Multiplatform.
- **Reliability and fault-tolerance**: Handle failures with built-in retries and restore the agent state at specific points during execution with the agent persistence feature.
- **Intelligent history compression**: Optimize token usage while maintaining context in long-running conversations using advanced built-in history compression techniques.


### Source-backed notes

- README includes Maven Central badge and dependency instructions for Gradle (Kotlin DSL), Gradle (Groovy), and Maven.
- README highlights Model Context Protocol integration as a feature area.
- Repo is Apache-2.0 licensed (verified via GitHub API).

### FAQ

- **How do I add Koog to a Kotlin project?**: Use the Gradle/Maven dependency snippets in README (e.g., `ai.koog:koog-agents:0.7.3`).
- **Is MCP supported?**: README explicitly mentions Model Context Protocol integration for using MCP tools in agents.
- **What license is Koog under?**: Apache-2.0 (verified via GitHub license metadata).

## Source & Thanks

> Created by [JetBrains](https://github.com/JetBrains). Licensed under Apache-2.0.
>
> [JetBrains/koog](https://github.com/JetBrains/koog) — ⭐ 4187

Thanks to the upstream maintainers and contributors for publishing this work under an open license.

---

<!-- ZH -->

## Quick Use

```kotlin
// Gradle (Kotlin DSL) from README
dependencies {
  implementation("ai.koog:koog-agents:0.7.3")
}
```

## Intro

Koog 是 JVM（Kotlin/Java）agent 框架，强调可预测与容错，并提供 MCP 工具集成入口，适合把工具链接入做成结构化接口；已验证 4187★，更新于 2026-05-14。

**Best for:** 用 Kotlin/Java 构建生产级 agent，并希望集成更结构化工具链的团队

**Works with:** Kotlin/Java 工程（Gradle/Maven）；README 介绍了 MCP 工具集成

**Setup time:** 12-25 minutes

### Key facts (verified)

- GitHub：4187 stars · 403 forks；最近更新 2026-05-14。
- 许可证：Apache-2.0；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中可对照的入口：`ai.koog:koog-agents:0.7.3`。

## Main

- 以库方式接入而不是复制模板：先加依赖，再逐步扩展 agent 能力面。

- 把 MCP 作为工具边界：结构化的工具接口让 agent 逻辑更易测试与复用。

- 优先保证可预测控制流：外部工具调用要把重试/超时当一等公民。

- 把最小可运行示例写进仓库：依赖 + demo 往往比抽象架构说明更能提升上手速度。

### README (excerpt)

**Koog**

[![Kotlin Beta](https://kotl.in/badges/beta.svg)](https://kotlinlang.org/docs/components-stability.html)
[![Maven Central](https://img.shields.io/maven-central/v/ai.koog/koog-agents)](https://search.maven.org/artifact/ai.koog/koog-agents)
[![JetBrains incubator project](https://jb.gg/badges/incubator.svg)](https://github.com/JetBrains#jetbrains-on-github)
[![Kotlin](https://img.shields.io/badge/kotlin-2.2-blue.svg?logo=kotlin)](http://kotlinlang.org)
[![CI status](https://img.shields.io/github/checks-status/JetBrains/koog/main)](https://github.com/JetBrains/koog/actions?query=branch%3Amain)
[![GitHub license](https://img.shields.io/github/license/JetBrains/koog)](LICENSE.txt)

Build status:

[![Checks](https://github.com/JetBrains/koog/actions/workflows/checks.yml/badge.svg?branch=develop)](https://github.com/JetBrains/koog/actions/workflows/checks.yml?query=branch%3Adevelop)
[![Heavy Tests](https://github.com/JetBrains/koog/actions/workflows/heavy-tests.yml/badge.svg?branch=develop)](https://github.com/JetBrains/koog/actions/workflows/heavy-tests.yml?query=branch%3Adevelop)
[![Ollama Tests](https://github.com/JetBrains/koog/actions/workflows/ollama-tests.yml/badge.svg?branch=develop)](https://github.com/JetBrains/koog/actions/workflows/ollama-tests.yml?query=branch%3Adevelop)

Useful links:

* [Documentation](https://docs.koog.ai/)
* [API reference](https://api.koog.ai/)
* [Slack channel](https://docs.koog.ai/koog-slack-channel/)
* [Issue tracker](https://youtrack.jetbrains.com/issues/KG)

## Overview

Koog is a Kotlin-based framework designed to build and run AI agents entirely in idiomatic Kotlin and Java API. It lets you create agents that can interact with tools, handle complex workflows, and communicate with users.

### Key features

Key features of Koog include:

- **Multiplatform development**: Deploy agents across JVM, JS, WasmJS, Android, and iOS targets using Kotlin Multiplatform.
- **Reliability and fault-tolerance**: Handle failures with built-in retries and restore the agent state at specific points during execution with the agent persistence feature.
- **Intelligent history compression**: Optimize token usage while maintaining context in long-running conversations using advanced built-in history compression techniques.


### Source-backed notes

- README 提供 Maven Central 标记，并给出 Gradle（Kotlin DSL / Groovy）与 Maven 的依赖配置示例。
- README 将 Model Context Protocol（MCP）集成列为功能点之一。
- 仓库为 Apache-2.0 许可证（已通过 GitHub API 复核）。

### FAQ

- **怎么在 Kotlin 工程里接入？**：按 README 使用 Gradle/Maven 依赖示例（例如 `ai.koog:koog-agents:0.7.3`）。
- **支持 MCP 吗？**：支持：README 明确提到 Model Context Protocol 集成，用于在 agent 中使用 MCP 工具。
- **Koog 的许可证是什么？**：Apache-2.0（已通过 GitHub 许可证元数据复核）。

## Source & Thanks

> Created by [JetBrains](https://github.com/JetBrains). Licensed under Apache-2.0.
>
> [JetBrains/koog](https://github.com/JetBrains/koog) — ⭐ 4187


---
Source: https://tokrepo.com/en/workflows/koog-kotlin-jvm-agent-framework-with-mcp-tools
Author: Agent Toolkit