快速入門

這份指南將協助您在專案中開始使用 Koog。

先決條件

--8<-- "quickstart-snippets.md:prerequisites"

安裝 Koog

--8<-- "quickstart-snippets.md:dependencies"

??? tip "Nightly 版本"

來自 develop 分支的 Nightly 版本會發佈到 [JetBrains Grazie Maven](https://packages.jetbrains.team/maven/p/grazi/grazie-platform-public) 存儲庫。

若要使用 Nightly 版本，請將以下存儲庫新增到您的組建組態：
`https://packages.jetbrains.team/maven/p/grazi/grazie-platform-public`。

接著將您的 Koog 相依性更新為所需的 Nightly 版本。Nightly 版本遵循以下模式：
`[next-major-version]-develop-[date]-[time]`。

您可以在[這裡](https://packages.jetbrains.team/maven/p/grazi/grazie-platform-public/ai/koog/koog-agents/)瀏覽可用的 Nightly 版本。

設定 API 金鑰

Koog 需要來自受支援的 LLM 提供者的 API 金鑰，或是本機執行的 LLM。

WARNING

避免在原始碼中硬編碼 API 金鑰。 請使用環境變數來儲存 API 金鑰。

=== "OpenAI"

取得您的 [OpenAI API 金鑰](https://platform.openai.com/api-keys) 並將其指派給 `OPENAI_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export OPENAI_API_KEY=your-api-key
    ```

=== "Windows"

    ```cmd
    setx OPENAI_API_KEY "your-api-key"
    ```

=== "Anthropic"

取得您的 [Anthropic API 金鑰](https://console.anthropic.com/settings/keys) 並將其指派給 `ANTHROPIC_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export ANTHROPIC_API_KEY=your-api-key
    ```

=== "Windows"

    ```cmd
    setx ANTHROPIC_API_KEY "your-api-key"
    ```

=== "Google"

取得您的 [Gemini API 金鑰](https://aistudio.google.com/app/api-keys) 並將其指派給 `GOOGLE_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export GOOGLE_API_KEY=your-api-key
    ```

=== "Windows"

    ```cmd
    setx GOOGLE_API_KEY "your-api-key"
    ```  

=== "DeepSeek"

取得您的 [DeepSeek API 金鑰](https://platform.deepseek.com/api_keys) 並將其指派給 `DEEPSEEK_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export DEEPSEEK_API_KEY=your-api-key
    ```

=== "Windows"

    ```cmd
    setx DEEPSEEK_API_KEY "your-api-key"
    ``` 

=== "OpenRouter"

取得您的 [OpenRouter API 金鑰](https://openrouter.ai/keys) 並將其指派給 `OPENROUTER_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export OPENROUTER_API_KEY=your-api-key
    ```

=== "Windows"

    ```cmd
    setx OPENROUTER_API_KEY "your-api-key"
    ```  

=== "Bedrock"

[產生 Amazon Bedrock API 金鑰](https://docs.aws.amazon.com/bedrock/latest/userguide/api-keys.html) 並將其指派給 `BEDROCK_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export BEDROCK_API_KEY=your-api-key
    ``` 

=== "Windows"

    ```cmd
    setx BEDROCK_API_KEY "your-api-key"
    ```  

=== "Mistral"

取得您的 [Mistral API 金鑰](https://console.mistral.ai/api-keys) 並將其指派給 `MISTRAL_API_KEY` 環境變數。

=== "Linux/macOS"

    ```shell
    export MISTRAL_API_KEY=your-api-key
    ``` 

=== "Windows"

    ```cmd
    setx MISTRAL_API_KEY "your-api-key"
    ``` 
    <!--- KNIT example-getting-started-01.txt -->

=== "Ollama"

按照 [Ollama 文件](https://docs.ollama.com/quickstart)中的說明，在 Ollama 中執行本機 LLM。

建立您的第一個 Koog Agent

=== "OpenAI"

以下範例透過 OpenAI API，使用 [`GPT-4o`](https://platform.openai.com/docs/models/gpt-4o) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleOpenAIExecutor
    import ai.koog.prompt.executor.clients.openai.OpenAIModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 OPENAI_API_KEY 環境變數取得 OpenAI API 金鑰
        val apiKey = System.getenv("OPENAI_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleOpenAIExecutor(apiKey),
            llmModel = OpenAIModels.Chat.GPT4o
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-01.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->
    ```java
    // 從 OPENAI_API_KEY 環境變數取得 OpenAI API 金鑰
    String apiKey = System.getenv("OPENAI_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleOpenAIExecutor(apiKey))
        .llmModel(OpenAIModels.Chat.GPT4o)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-01.java -->

該範例可能會產生以下輸出：

```
Hello! I'm here to help you with whatever you need. Here are just a few things I can do:

- Answer questions.
- Explain concepts or topics you're curious about.
- Provide step-by-step instructions for tasks.
- Offer advice, notes, or ideas.
- Help with research or summarize complex material.
- Write or edit text, emails, or other documents.
- Brainstorm creative projects or solutions.
- Solve problems or calculations.

Let me know what you need help with—I’m here for you!
```
<!--- KNIT example-getting-started-02.txt -->

=== "Anthropic"

以下範例透過 Anthropic API，使用 [`Claude Opus 4.1`](https://www.anthropic.com/news/claude-opus-4-1) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleAnthropicExecutor
    import ai.koog.prompt.executor.clients.anthropic.AnthropicModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 ANTHROPIC_API_KEY 環境變數取得 Anthropic API 金鑰
        val apiKey = System.getenv("ANTHROPIC_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleAnthropicExecutor(apiKey),
            llmModel = AnthropicModels.Opus_4_1
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-02.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->
    ```java
    // 從 ANTHROPIC_API_KEY 環境變數取得 Anthropic API 金鑰
    String apiKey = System.getenv("ANTHROPIC_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleAnthropicExecutor(apiKey))
        .llmModel(AnthropicModels.Opus_4_1)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-02.java -->

該範例可能會產生以下輸出：

```
Hello! I can help you with:

- **Answering questions** and explaining topics
- **Writing** - drafting, editing, proofreading
- **Learning** - homework, math, study help
- **Problem-solving** and brainstorming
- **Research** and information finding
- **General tasks** - instructions, planning, recommendations

What do you need help with today?
```
<!--- KNIT example-getting-started-03.txt -->

=== "Google"

以下範例透過 Gemini API，使用 [`Gemini 2.5 Pro`](https://cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-pro) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleGoogleAIExecutor
    import ai.koog.prompt.executor.clients.google.GoogleModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 GOOGLE_API_KEY 環境變數取得 Gemini API 金鑰
        val apiKey = System.getenv("GOOGLE_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleGoogleAIExecutor(apiKey),
            llmModel = GoogleModels.Gemini2_5Pro
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-03.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->
    ```java
    // 從 GOOGLE_API_KEY 環境變數取得 Gemini API 金鑰
    String apiKey = System.getenv("GOOGLE_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleGoogleAIExecutor(apiKey))
        .llmModel(GoogleModels.Gemini2_5Pro)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-03.java -->

該範例可能會產生以下輸出：

```
I'm an AI that can help you with tasks involving language and information. You can ask me to:

*   **Answer questions**
*   **Write or edit text** (emails, stories, code, etc.)
*   **Brainstorm ideas**
*   **Summarize long documents**
*   **Plan things** (like trips or projects)
*   **Be a creative partner**

Just tell me what you need
```
<!--- KNIT example-getting-started-04.txt -->

=== "DeepSeek"

以下範例透過 DeepSeek API，使用 `deepseek-v4-flash` 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.clients.deepseek.DeepSeekLLMClient
    import ai.koog.prompt.executor.llms.MultiLLMPromptExecutor
    import ai.koog.prompt.executor.clients.deepseek.DeepSeekModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 DEEPSEEK_API_KEY 環境變數取得 DeepSeek API 金鑰
        val apiKey = System.getenv("DEEPSEEK_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 LLM 用戶端
        val deepSeekClient = DeepSeekLLMClient(apiKey)
    
        // 建立 Agent
        val agent = AIAgent(
            // 使用 LLM 用戶端建立提示詞執行器
            promptExecutor = MultiLLMPromptExecutor(deepSeekClient),
            // 提供模型
            llmModel = DeepSeekModels.DeepSeekV4Flash
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-04.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->
    ```java
    // 從 DEEPSEEK_API_KEY 環境變數取得 DeepSeek API 金鑰
    String apiKey = System.getenv("DEEPSEEK_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 LLM 用戶端
    DeepSeekLLMClient deepSeekClient = new DeepSeekLLMClient(apiKey);

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        // 使用 LLM 用戶端建立提示詞執行器
        .promptExecutor(new MultiLLMPromptExecutor(deepSeekClient))
        // 提供模型
        .llmModel(DeepSeekModels.DeepSeekV4Flash)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-04.java -->

該範例可能會產生以下輸出：

```
Hello! I'm here to assist you with a wide range of tasks, including answering questions, providing information, helping with problem-solving, offering creative ideas, and even just chatting. Whether you need help with research, writing, learning something new, or simply want to discuss a topic, feel free to ask—I’m happy to help! 😊
```
<!--- KNIT example-getting-started-05.txt -->

=== "OpenRouter"

以下範例透過 OpenRouter API，使用 [`GPT-4o`](https://openrouter.ai/openai/gpt-4o) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleOpenRouterExecutor
    import ai.koog.prompt.executor.clients.openrouter.OpenRouterModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 OPENROUTER_API_KEY 環境變數取得 OpenRouter API 金鑰
        val apiKey = System.getenv("OPENROUTER_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleOpenRouterExecutor(apiKey),
            llmModel = OpenRouterModels.GPT4o
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-05.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->
    ```java
    // 從 OPENROUTER_API_KEY 環境變數取得 OpenRouter API 金鑰
    String apiKey = System.getenv("OPENROUTER_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleOpenRouterExecutor(apiKey))
        .llmModel(OpenRouterModels.GPT4o)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-05.java -->

該範例可能會產生以下輸出：

```
I can answer questions, help with writing, solve problems, organize tasks, and more—just let me know what you need!
```
<!--- KNIT example-getting-started-06.txt -->

=== "Bedrock"

以下範例透過 Bedrock API，使用 [`Claude Sonnet 4.5`](https://www.anthropic.com/news/claude-sonnet-4-5) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleBedrockExecutorWithBearerToken
    import ai.koog.prompt.executor.clients.bedrock.BedrockModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 BEDROCK_API_KEY 環境變數取得 Bedrock API 金鑰
        val apiKey = System.getenv("BEDROCK_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleBedrockExecutorWithBearerToken(apiKey),
            llmModel = BedrockModels.AnthropicClaude4_5Sonnet
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-06.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->
    ```java
    // 從 BEDROCK_API_KEY 環境變數取得 Bedrock API 金鑰
    String apiKey = System.getenv("BEDROCK_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleBedrockExecutorWithBearerToken(apiKey, new BedrockClientSettings()))
        .llmModel(BedrockModels.INSTANCE.getAnthropicClaude4_5Sonnet())
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-06.java -->

該範例可能會產生以下輸出：

```
Hello! I'm a helpful assistant and I can assist you in many ways, including:

- **Answering questions** on a wide range of topics (science, history, technology, etc.)
- **Writing help** - drafting emails, essays, creative content, or editing text
- **Problem-solving** - working through math problems, logic puzzles, or troubleshooting issues
- **Learning support** - explaining concepts, providing study notes, or tutoring
- **Planning & organizing** - helping with projects, schedules, or breaking down tasks
- **Coding assistance** - explaining programming concepts or helping debug code
- **Creative brainstorming** - generating ideas for projects, stories, or solutions
- **General conversation** - discussing topics or just chatting

 What would you like help with today?
```
<!--- KNIT example-getting-started-07.txt -->

=== "Mistral"

以下範例透過 Mistral AI API，使用 [`Mistral Medium 3.1`](https://docs.mistral.ai/models/mistral-medium-3-1-25-08) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleMistralAIExecutor
    import ai.koog.prompt.executor.clients.mistralai.MistralAIModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 從 MISTRAL_API_KEY 環境變數取得 Mistral AI API 金鑰
        val apiKey = System.getenv("MISTRAL_API_KEY")
            ?: error("The API key is not set.")
        
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleMistralAIExecutor(apiKey),
            llmModel = MistralAIModels.Chat.MistralMedium31
        )
    
        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-07.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->   
    ```java
    // 從 MISTRAL_API_KEY 環境變數取得 Mistral AI API 金鑰
    String apiKey = System.getenv("MISTRAL_API_KEY");
    if (apiKey == null) {
        throw new RuntimeException("The API key is not set.");
    }

    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleMistralAIExecutor(apiKey))
        .llmModel(MistralAIModels.Chat.MistralMedium31)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-07.java -->

該範例可能會產生以下輸出：

```
I can assist you with a wide range of topics and tasks. Here are some examples:

1. **Answering questions**: I can provide information on various subjects, including history, science, technology, literature, and more.
2. **Providing definitions**: If you're unsure about the meaning of a word or phrase, I can help define it for you.
3. **Generating text**: Whether it's writing an email, creating content for social media, or composing a story, I can help with text generation.
4. **Translation**: I can translate text from one language to another.
5. **Conversation**: We can have a chat about any topic that interests you, and I'll respond accordingly.
6. **Language practice**: If you're learning a new language, I can help with pronunciation, grammar, and vocabulary practice.
7. **Brainstorming**: If you're stuck on a problem or need ideas for a project, I can help brainstorm solutions.
8. **Summarization**: If you have a long piece of text and want a summary, I can condense it for you.

What's on your mind? Is there something specific you'd like help with?
```
<!--- KNIT example-getting-started-08.txt -->

=== "Ollama"

以下範例使用經由 Ollama 本機執行的 [`llama3.2`](https://ollama.com/library/llama3.2) 模型建立並執行一個簡單的 Koog Agent。

=== "Kotlin"

    <!--- INCLUDE
    import ai.koog.agents.core.agent.AIAgent
    import ai.koog.prompt.executor.llms.all.simpleOllamaAIExecutor
    import ai.koog.prompt.executor.ollama.client.OllamaModels
    import kotlinx.coroutines.runBlocking
    -->
    ```kotlin
    fun main() = runBlocking {
        // 建立 Agent
        val agent = AIAgent(
            promptExecutor = simpleOllamaAIExecutor(),
            llmModel = OllamaModels.Meta.LLAMA_3_2
        )

        // 執行 Agent
        val result = agent.run("Hello! How can you help me?")
        println(result)
    }
    ```
    <!--- KNIT example-getting-started-08.kt -->

=== "Java"

    <!--- INCLUDE
    /**
    -->
    <!--- SUFFIX
    **/
    -->  
    ```java
    // 建立 Agent
    AIAgent<String, String> agent = AIAgent.builder()
        .promptExecutor(simpleOllamaAIExecutor("http://localhost:11434"))
        .llmModel(OllamaModels.Meta.LLAMA_3_2)
        .build();

    // 執行 Agent
    String result = agent.run("Hello! How can you help me?");
    System.out.println(result);
    ```
    <!--- KNIT example-getting-started-java-08.java -->

該範例可能會產生以下輸出：

```
I can assist with various tasks such as answering questions, providing information, and even helping with language-related tasks like proofreading or writing suggestions. What's on your mind today?
```
<!--- KNIT example-getting-started-09.txt -->

後續步驟

• 進一步了解 Agent 類型