### Basic Setup with Claude Subagents

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/TaskTools.md

Configure the Task tool with Claude subagents and build the main chat client. This example demonstrates how to use the agent for natural language queries that delegate to sub-agents.

```java
import org.springaicommunity.agent.tools.task.TaskTool;
import org.springaicommunity.agent.tools.task.claude.ClaudeSubagentType;

@Configuration
public class AgentConfig {

    @Bean
    CommandLineRunner demo(ChatClient.Builder chatClientBuilder) {
        return args -> {
            // Configure Task tool with Claude subagents
            var taskTool = TaskTool.builder()
                .subagentTypes(ClaudeSubagentType.builder()
                    .chatClientBuilder("default", chatClientBuilder)
                    .build())
                .build();

            // Build main chat client with Task tool
            ChatClient chatClient = chatClientBuilder
                .defaultToolCallbacks(taskTool)
                .build();

            // Use naturally - agent will delegate to sub-agents
            String response = chatClient
                .prompt("Explore the authentication module and explain how it works")
                .call()
                .content();
        };
    }
}
```

--------------------------------

### Example Workflow with ShellTools

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/ShellTools.md

Demonstrates a typical workflow: starting a long-running background task, periodically checking its progress, and finally killing the process. Includes starting a process, getting its output with filtering, and then terminating it.

```java
ShellTools shellTools = new ShellTools();

// 1. Start a long-running background task
String start = shellTools.bash(
    "python train_model.py",
    null,
    "Train ML model",
    true
);
// Returns: "bash_id: shell_1234567890 ..."

// 2. Periodically check progress
String output1 = shellTools.bashOutput("shell_1234567890", ".*epoch.*\n");
// Shows only lines containing "epoch"

Thread.sleep(5000);

String output2 = shellTools.bashOutput("shell_1234567890", null);
// Shows all new output since last check

// 3. Kill if needed
String killResult = shellTools.killShell("shell_1234567890");
// Returns: "Successfully killed shell: shell_1234567890"
```

--------------------------------

### Manual MemoryTools Setup vs. Advisor Setup

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/AutoMemoryToolsAdvisor.md

Compares the manual setup of AutoMemoryTools with the simplified setup using AutoMemoryToolsAdvisor. The advisor reduces boilerplate by handling prompt injection and tool registration automatically.

```java
// Manual setup
AutoMemoryTools memoryTools = AutoMemoryTools.builder()
    .memoriesDir("/path/to/memories")
    .build();

String memoryPrompt = new PromptTemplate(memorySystemPromptResource)
    .render(Map.of("MEMORIES_ROOT_DIERCTORY", memoriesDir));

ChatClient chatClient = ChatClient.builder(chatModel)
    .defaultSystem(baseSystemPrompt + "\n\n" + memoryPrompt)
    .defaultTools(memoryTools)
    .build();
```

```java
// Advisor setup — equivalent, less boilerplate
ChatClient chatClient = ChatClient.builder(chatModel)
    .defaultAdvisors(
        AutoAutoMemoryToolsAdvisor.builder()
            .memoriesRootDirectory("/path/to/memories")
            .build())
    .build();
```

--------------------------------

### Full Shell Workflow Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/ShellTools.md

Demonstrates a complete workflow: starting a long-running background task, periodically checking its output, and finally killing the process. Includes simulated progress checks and output retrieval.

```java
ShellTools shellTools = new ShellTools();

// 1. Start a long-running background task
String start = shellTools.bash(
    "python train_model.py",
    null,
    "Train ML model",
    true
);
// Returns: "bash_id: shell_1234567890 ..."

// 2. Periodically check progress
String output1 = shellTools.bashOutput("shell_1234567890", ".*epoch.*");
// Shows only lines containing "epoch"

Thread.sleep(5000);

String output2 = shellTools.bashOutput("shell_1234567890", null);
// Shows all new output since last check

// 3. Kill if needed
String killResult = shellTools.killShell("shell_1234567890");
// Returns: "Successfully killed shell: shell_1234567890"
```

--------------------------------

### Data Analysis Skill Setup

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

Instructions for setting up the Data Analysis skill by installing its Python dependencies.

```bash
cd .claude/skills/data-analysis
pip install -r requirements.txt
```

--------------------------------

### Build and Run Project

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/index.md

Commands to build the entire project using Maven and run a specific example application. Ensure you have Java 17+, Spring Boot 3.x/4.x, Spring AI 2.0.0+, and Maven 3.6+ installed.

```bash
# Build the entire project
mvn clean install

# Run an example
cd examples/code-agent-demo  # or examples/skills-demo
mvn spring-boot:run
```

--------------------------------

### Run the Demo

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/memory/memory-filesystem-tools-demo/README.md

Execute the demo using Maven to start the Spring Boot application.

```bash
mvn spring-boot:run
```

--------------------------------

### SmartWebFetchTool Builder with All Options

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SmartWebFetchTool.md

Example of configuring the SmartWebFetchTool builder with all available optional parameters for custom behavior.

```java
SmartWebFetchTool webFetch = SmartWebFetchTool.builder(chatClient)
    .maxContentLength(150_000)           // Process up to 150KB
    .domainSafetyCheck(true)             // Check domain safety
    .failOpenOnSafetyCheckError(true)    // Allow fetch if safety check errors
    .maxCacheSize(200)                   // Cache up to 200 entries
    .maxRetries(3)                       // Retry up to 3 times
    .build();
```

--------------------------------

### Install Dependencies with uv

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/subagent-a2a-demo/src/main/resources/skills/ai-tutor/README.md

Install the project dependencies using the uv package manager.

```bash
uv sync
```

--------------------------------

### Quick Start with TodoWriteTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/TodoWriteTool.md

Initialize a ChatClient with the TodoWriteTool and necessary advisors. Ensure Chat Memory and ToolCallAdvisor are configured for optimal performance.

```java
ChatClient chatClient = chatClientBuilder
    .defaultTools(TodoWriteTool.builder().build())
    .defaultAdvisors(
        ToolCallAdvisor.builder().conversationHistoryEnabled(false).build(),
        MessageChatMemoryAdvisor.builder(MessageWindowChatMemory.builder().build()).build())
    .build();
```

--------------------------------

### SKILL.md Format Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

The SKILL.md file includes YAML front-matter with metadata and Markdown content for instructions and examples.

```markdown
---
name: my-skill
description: Comprehensive description including what the skill does,
  when to use it, and trigger keywords users might mention.
allowed-tools: Read, Grep, Bash
model: claude-sonnet-4-5-20250929
---

# Skill Title

## Overview
Brief introduction to what this skill does.

## Instructions
Step-by-step guidance for the AI agent.

## Examples
Concrete examples demonstrating skill usage.

## Additional Resources
- For complete details, see [reference.md](reference.md)
- For usage examples, see [examples.md](examples.md)
```

--------------------------------

### Example Prompt for TodoWriteTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/todo-demo/README.md

An example prompt demonstrating a complex request that the agent should decompose into trackable todo items using the TodoWrite tool.

```plaintext
Find the top 10 Tom Hanks movies, then group them in groups of 2,
and finally print the title name inverted (e.g. last char first).
Use TodoWrite to organize your tasks.
```

--------------------------------

### Example Conversation - Session 2

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/memory/memory-tools-demo/README.md

Run the demo again and ask the agent about stored information to see its recall capability.

```bash
USER> What do you know about me?
ASSISTANT> You're Alice, a backend engineer. You prefer short answers.
           You're migrating from PostgreSQL to CockroachDB this quarter.
```

--------------------------------

### Skill File Structure Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

A typical skill directory includes the required SKILL.md file and optional supporting files like reference documents, examples, and scripts.

```text
examples/.claude/skills/
└── my-skill/
    ├── SKILL.md          # Required: Skill definition
    ├── reference.md      # Optional: Detailed documentation
    ├── examples.md       # Optional: Usage examples
    ├── scripts/          # Optional: Helper scripts
    │   └── process.py
    └── pyproject.toml    # Optional: Python dependencies
```

--------------------------------

### Basic AskUserQuestionTool Setup

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/AskUserQuestionTool.md

Demonstrates how to build and configure the AskUserQuestionTool with a custom question handler and integrate it with a ChatClient. Ensure your QuestionHandler is thread-safe if it maintains shared state.

```java
AskUserQuestionTool askTool = AskUserQuestionTool.builder()
    .questionHandler(questions -> {
        // Display questions to user via your UI and collect answers
        return collectUserAnswers(questions);
    })
    .build();

ChatClient chatClient = chatClientBuilder
    .defaultTools(askTool)
    .build();
```

--------------------------------

### Start Shell with Description

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/ShellTools.md

Starts a shell process and assigns a descriptive name for better logging. This is useful for identifying the purpose of background tasks.

```java
shellTools.bash("git status", null, "Check git status", null);
```

--------------------------------

### Skill Examples (Bash and Python)

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

Illustrative examples of how skills can be invoked using bash scripts or Python code, demonstrating common tasks like text extraction, merging PDFs, and filling forms.

```bash
python scripts/extract_text.py input.pdf output.txt
```

```bash
python scripts/merge.py file1.pdf file2.pdf output.pdf
```

```python
from pdf_processor import fill_form

fill_form('template.pdf', {
    'name': 'John Doe',
    'date': '2025-01-04'
}, 'filled.pdf')
```

--------------------------------

### Find All Configuration Files

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Example of finding all '.yml' files within the 'src/main/resources' directory recursively.

```java
String configs = globTool.glob("**/*.yml", "./src/main/resources");
```

--------------------------------

### Build Project with Maven

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/README.md

Use this command to build the project from the root directory. Ensure Java 17+ and Maven 3.6+ are installed.

```bash
mvn clean install
```

--------------------------------

### Agent Configuration for Model Routing (Fast/Cheap)

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/TaskTools.md

Example of configuring an agent to use a faster and cheaper model like 'haiku' for simple tasks.

```markdown
---
name: quick-search
model: haiku
---

```

--------------------------------

### Example Conversation - Session 1

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/memory/memory-tools-demo/README.md

Interact with the agent in the first session, providing information and preferences.

```bash
USER> My name is Alice. I'm a backend engineer and I prefer short answers.
USER> Remember that we're migrating from PostgreSQL to CockroachDB this quarter.
USER> exit
```

--------------------------------

### Start Background Task with Description

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/ShellTools.md

Starts a shell command to run in the background. Use this for commands expected to take longer than 30 seconds. A descriptive name aids in logging and management.

```java
shellTools.bash("git status", null, "Check git status", null);
```

```java
shellTools.bash("npm run test", null, "Run test suite", true);
```

--------------------------------

### Register SkillsTool with Skill Directories

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/README.md

Demonstrates registering SkillsTool with specified skill directories and other necessary tools for AI agent setup.

```java
// Register SkillsTool with skill directories
ChatClient chatClient = chatClientBuilder
    .defaultToolCallbacks(SkillsTool.builder()
        .addSkillsDirectory(".claude/skills")
        .build())
    .defaultTools(FileSystemTools.builder().build())  // For reading reference files
    .defaultTools(new ShellTools())       // For executing scripts
    .build();
```

--------------------------------

### Quick Start AutoAutoMemoryToolsAdvisor

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/AutoMemoryToolsAdvisor.md

Initialize AutoAutoMemoryToolsAdvisor with a memories root directory and integrate it into a ChatClient.

```java
AutoAutoMemoryToolsAdvisor memoryAdvisor = AutoAutoMemoryToolsAdvisor.builder()
    .memoriesRootDirectory("/home/user/.agent/memories")
    .build();

ChatClient chatClient = ChatClient.builder(chatModel)
    .defaultAdvisors(memoryAdvisor)
    .build();
```

--------------------------------

### Find All Java Source Files

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Example of finding all '.java' files within the 'src/main' directory recursively.

```java
String javaSources = globTool.glob("**/*.java", "./src/main");
```

--------------------------------

### Configure Skills Directory

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/skills-demo/README.md

Configure the application properties to specify the directory where skills are loaded from. This example uses a classpath resource.

```properties
agent.skills.dirs=classpath:/.claude/skills
```

--------------------------------

### Example QuestionHandler Implementation

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/AskUserQuestionTool.md

A sample implementation of the QuestionHandler interface that iterates through questions, prompts the user for an answer to each, and returns a map of questions to answers.

```java
QuestionHandler handler = questions -> {
    Map<String, String> answers = new HashMap<>();
    for (Question q : questions) {
        String answer = promptUser(q);
        answers.put(q.question(), answer);
    }
    return answers;
};
```

--------------------------------

### Match Files Starting with 'Test'

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Find all files that begin with the prefix 'Test' and have the '.java' extension within the specified directory.

```java
String testFiles = globTool.glob("Test*.java", "./src");
```

--------------------------------

### Spring Boot Configuration for SmartWebFetchTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SmartWebFetchTool.md

Example of configuring SmartWebFetchTool as a Spring Bean, setting various parameters like maxContentLength, domainSafetyCheck, and maxRetries.

```java
@Configuration
public class ToolsConfig {

    @Bean
    public SmartWebFetchTool smartWebFetchTool(ChatClient.Builder chatClientBuilder) {
        ChatClient chatClient = chatClientBuilder.build();

        return SmartWebFetchTool.builder(chatClient)
            .maxContentLength(150_000)
            .domainSafetyCheck(true)
            .failOpenOnSafetyCheckError(true)
            .maxCacheSize(100)
            .maxRetries(2)
            .build();
    }
}
```

--------------------------------

### Skill Execution Examples - Python Script

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/SkillsTool.md

Demonstrates calling a Python script to merge multiple PDF files. Ensure the script and its dependencies are accessible.

```bash
python scripts/merge.py file1.pdf file2.pdf output.pdf
```

--------------------------------

### Document a REST Endpoint

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

Example of documenting a GET endpoint with parameters, response format, and error codes in Markdown.

```markdown
## GET /api/users/{id}

Retrieve user by ID.

**Parameters:**
- `id` (path, required): User ID

**Response:** `200 OK`

```json
{
  "id": 1,
  "name": "John Doe",
  "email": "john@example.com"
}
```

**Errors:**
- `404`: User not found
```

--------------------------------

### Custom Sub-Agent File Structure Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/TaskTools.md

Illustrates the directory structure for custom sub-agents, showing how Markdown files with YAML front matter are organized within a project.

```markdown
project-root/
├── .claude/
│   └── agents/
│       ├── code-reviewer.md
│       ├── test-runner.md
│       └── spring-ai-expert.md
```

--------------------------------

### AgentEnvironment.gitStatus() Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/AgentEnvironment.md

Collects and prints the git repository status, including branch, main branch, modified files, and recent commits. Returns an empty string if git is not installed or the directory is not a git repo.

```java
String gitStatus = AgentEnvironment.gitStatus();
System.out.println(gitStatus);
// gitStatus: This is the git status at the start of the conversation...
// Current branch: feature-branch
// Main branch (you will usually use this for PRs): main
// Status:
// M README.md
// ...
```

--------------------------------

### Technology Stack Research with Trusted Sources

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/BraveWebSearchTool.md

Find tutorials or guides from a curated list of trusted sources. This ensures the information is relevant and from reputable platforms.

```java
// Find tutorials from trusted sources
String tutorials = searchTool.webSearch(
    "Spring AI LangChain4j integration tutorial",
    List.of("spring.io", "github.com", "medium.com", "dev.to"),
    null
);
```

--------------------------------

### Build and Run JAR

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/ask-user-question-demo/README.md

Package the application into a JAR file and execute it.

```bash
mvn clean package
java -jar target/chat-demo2-0.1.0-SNAPSHOT.jar
```

--------------------------------

### Memory Write Operation Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/memory/memory-filesystem-tools-demo/README.md

Illustrates the format of a typed memory file (.md) with YAML frontmatter, including name, description, and type, followed by the memory content.

```markdown
---
name: user profile
description: Alice — backend engineer, prefers short answers
type: user
---

Backend engineer named Alice.
Prefers concise, direct responses without trailing summaries.
```

--------------------------------

### Clone Repository and Navigate

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/subagent-a2a-demo/src/main/resources/skills/ai-tutor/README.md

Clone the repository containing the AI Tutor Skill and navigate into the project directory.

```bash
git clone <repository-url>
cd ai-tutor-skill
```

--------------------------------

### Bad Skill Description Example (YAML)

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

Examples of vague skill descriptions in YAML format that should be avoided.

```yaml
# ❌ BAD
description: Database helper
```

```yaml
description: Works with files
```

--------------------------------

### Set API Key and Run Demo

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/todo-demo/README.md

Configure your environment with API keys for your chosen model provider and Brave search, then execute the Spring Boot application.

```bash
# Set API key for your chosen model provider
export GOOGLE_CLOUD_PROJECT=your-project-id
# or: export ANTHROPIC_API_KEY=your-key
# or: export OPENAI_API_KEY=your-key

# Optional: Enable web search
export BRAVE_API_KEY=your-brave-key

# Run
./mvnw spring-boot:run -pl examples/todo-demo
```

--------------------------------

### Set Environment Variables and Run

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/code-agent-demo/README.md

Set necessary API keys as environment variables and navigate to the demo directory. Then, run the Spring Boot application using Maven.

```bash
# 1. Set environment variables
export ANTHROPIC_API_KEY=your-key-here
export BRAVE_API_KEY=your-brave-key  # optional

# 2. Run
cd examples/code-agent-demo
mvn spring-boot:run
```

--------------------------------

### Skill Execution Examples - Bash Script

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/SkillsTool.md

Example of how to invoke a bash script for a specific task. This is useful for automating command-line operations.

```bash
python scripts/extract_text.py input.pdf output.txt
```

--------------------------------

### ChatClient Setup with Tools and Advisors

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/code-agent-demo/README.md

Configure the ChatClient in Application.java by setting the system prompt, tool callbacks for MCP and skills, and adding agent tools and advisors for memory and tool execution.

```java
ChatClient chatClient = chatClientBuilder
    .defaultSystem(systemPrompt)                        // MAIN_AGENT_SYSTEM_PROMPT_V2.md
    .defaultToolCallbacks(mcpToolCallbackProvider)      // MCP tools
    .defaultToolCallbacks(skillsTool)                   // Skills
    .defaultTools(AskUserQuestionTool, TodoWriteTool,   // Agent tools
                  ShellTools, FileSystemTools,
                  SmartWebFetchTool, BraveWebSearchTool, GrepTool)
    .defaultAdvisors(ToolCallAdvisor,                   // Tool execution
                     MessageChatMemoryAdvisor)          // 500-message memory
    .build();
```

--------------------------------

### Automatic Delegation Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/TaskTools.md

The main agent automatically delegates tasks to sub-agents based on their descriptions. This example shows a prompt that triggers exploration.

```java
String response = chatClient
    .prompt("How does authentication work in this codebase?")
    .call()
    .content();
```

--------------------------------

### Find Files Starting with a Pattern using GlobTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Use GlobTool to find files whose names start with a specific pattern, such as 'Test'.

```java
// Files starting with pattern
globTool.glob("**/Test*.java", "./src");
```

--------------------------------

### Build and Use BraveWebSearchTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/README.md

Demonstrates how to build a BraveWebSearchTool with an API key and perform web searches, including using search operators.

```java
// Build with API key
BraveWebSearchTool searchTool = BraveWebSearchTool.builder(apiKey)
    .resultCount(10)  // Optional: default 10
    .build();

// Search the web
String results = searchTool.webSearch(
    "Spring AI features 2025",
    null,  // allowedDomains (optional)
    null   // blockedDomains (optional)
);

// Or use search operators for efficiency
String results2 = searchTool.webSearch("Spring AI site:spring.io", null, null);
```

--------------------------------

### Complete TaskTool Example with Chat Client

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/TaskTools.md

Demonstrates setting up a TaskTool with Claude subagents and integrating it into a main chat client for interactive use. Includes configuration for subagents, tools, and advisors.

```java
import org.springframework.boot.CommandLineRunner;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.context.annotation.Bean;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;

import java.util.List;
import java.util.Scanner;

import ai.langchain.spring.core.ChatClient;
import ai.langchain.spring.core.ToolCallback;
import ai.langchain.spring.core.advisors.ToolCallAdvisor;
import ai.langchain.spring.core.advisors.MessageChatMemoryAdvisor;
import ai.langchain.spring.core.memory.MessageWindowChatMemory;
import ai.langchain.agents.task.TaskTool;
import ai.langchain.agents.task.ClaudeSubagentType;
import ai.langchain.agents.tools.FileSystemTools;
import ai.langchain.agents.tools.GrepTool;
import ai.langchain.agents.tools.GlobTool;
import ai.langchain.agents.tools.ShellTools;
import ai.langchain.agents.tools.TodoWriteTool;
import ai.langchain.agents.tools.ToolCallAdvisor;
import ai.langchain.agents.tools.MyLoggingAdvisor;

@SpringBootApplication
public class Application {

    @Bean
    CommandLineRunner demo(
            ChatClient.Builder chatClientBuilder,
            @Value("${agent.skills.paths}") List<Resource> skillPaths,
            @Value("${BRAVE_API_KEY:#{null}}") String braveApiKey) {

        return args -> {
            // Configure Task tool with Claude subagents
            var taskTool = TaskTool.builder()
                .subagentTypes(ClaudeSubagentType.builder()
                    .chatClientBuilder("default",
                        chatClientBuilder.clone().defaultAdvisors(new MyLoggingAdvisor(0, "[TASK]")))
                    .skillsResources(skillPaths)
                    .braveApiKey(braveApiKey)
                    .build())
                .build();

            // Build main chat client
            ChatClient chatClient = chatClientBuilder
                .defaultToolCallbacks(taskTool)
                .defaultTools(
                    FileSystemTools.builder().build(),
                    GrepTool.builder().build(),
                    GlobTool.builder().build(),
                    ShellTools.builder().build(),
                    TodoWriteTool.builder().build()
                )
                .defaultAdvisors(
                    ToolCallAdvisor.builder()
                        .conversationHistoryEnabled(false)
                        .build(),
                    MessageChatMemoryAdvisor.builder(
                        MessageWindowChatMemory.builder().maxMessages(500).build()
                    ).build()
                )
                .build();

            // Interactive chat loop
            try (Scanner scanner = new Scanner(System.in)) {
                while (true) {
                    System.out.print("\nUSER: ");
                    String response = chatClient
                        .prompt(scanner.nextLine())
                        .call()
                        .content();
                    System.out.println("\nASSISTANT: " + response);
                }
            }
        };
    }
}
```

--------------------------------

### Automatic Delegation Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/TaskTools.md

The main agent automatically delegates tasks to sub-agents based on their descriptions. This example shows a prompt that triggers automatic delegation.

```java
String response = chatClient
    .prompt("How does authentication work in this codebase?")
    .call()
    .content();
// Main agent recognizes this as exploration task and delegates to Explore sub-agent
```

--------------------------------

### Single-Select Question Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/AskUserQuestionTool.md

An example of a single-select question format that the AI agent can generate. The 'multiSelect' field is set to false, indicating that only one option can be chosen.

```json
{
    "question": "Which library should we use for date formatting?",
    "header": "Library",
    "options": [
        { "label": "Moment.js", "description": "Popular but large library" },
        { "label": "Day.js", "description": "Lightweight Moment.js alternative" },
        { "label": "date-fns", "description": "Modular and tree-shakeable" }
    ],
    "multiSelect": false
}
```

--------------------------------

### Basic SmartWebFetchTool Usage

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SmartWebFetchTool.md

Demonstrates how to build and use the SmartWebFetchTool to fetch and summarize web content. Requires a ChatClient instance.

```java
SmartWebFetchTool webFetch = SmartWebFetchTool.builder(chatClient).build();

String result = webFetch.webFetch(
    "https://docs.spring.io/spring-ai/reference/",
    "What are the key features of Spring AI?"
);

System.out.println(result);
```

--------------------------------

### Good Skill Description Example (YAML)

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

An example of a well-written skill description in YAML format, including specific capabilities and trigger keywords for better AI matching.

```yaml
# ✅ GOOD
description: Process Excel files: read sheets, analyze data, create charts,
  generate reports. Use when working with .xlsx, .xls files, spreadsheets,
  or when user mentions Excel, data analysis, or pivot tables.
```

```yaml
# ✅ GOOD
description: Analyze SQL databases, write optimized queries, explain execution
  plans. Use when user mentions: databases, SQL, MySQL, PostgreSQL, queries,
  tables, indexes, or performance tuning.
```

--------------------------------

### Default GrepTool Configuration and Usage

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GrepTool.md

Demonstrates how to create a GrepTool with default settings and perform a basic search for a pattern in the current directory. Uses `files_with_matches` output mode.

```java
GrepTool grepTool = GrepTool.builder().build();

String result = grepTool.grep(
    "TODO",              // pattern
    null,                // path (uses current directory)
    null,                // glob
    OutputMode.files_with_matches,  // outputMode
    null, null, null,    // context lines (before, after, surrounding)
    null,                // showLineNumbers
    null,                // caseInsensitive
    null,                // type
    null, null, null     // headLimit, offset, multiline
);
```

--------------------------------

### Skill Execution Examples - Python Function

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/SkillsTool.md

Example of using a Python function to fill a PDF form with provided data. This requires the 'pdf_processor' library and specific template files.

```python
from pdf_processor import fill_form

fill_form('template.pdf', {
    'name': 'John Doe',
    'date': '2025-01-04'
}, 'filled.pdf')
```

--------------------------------

### Build and Use TodoWriteTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/README.md

Demonstrates building a TodoWriteTool and managing a task list with TodoItem objects.

```java
TodoWriteTool todoTool = TodoWriteTool.builder().build();

// Create and manage task list
Todos todos = new Todos(List.of(
    new TodoItem("Read configuration", Status.completed, "Reading configuration"),
    new TodoItem("Parse settings", Status.in_progress, "Parsing settings"),
    new TodoItem("Validate config", Status.pending, "Validating config")
));

todoTool.todoWrite(todos);
```

--------------------------------

### Run Spring Boot Demo

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/ask-user-question-demo/README.md

Navigate to the demo directory and run the Spring Boot application using Maven.

```bash
cd examples/ask-user-question-demo
mvn spring-boot:run
```

--------------------------------

### SmartWebFetchTool Caching Behavior Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SmartWebFetchTool.md

Demonstrates how the tool uses URL and prompt to create cache keys. Different prompts for the same URL result in different cache entries, while identical URL and prompt reuse the cache.

```java
// These create DIFFERENT cache entries
webFetch.webFetch("https://example.com", "What is the main topic?");
webFetch.webFetch("https://example.com", "List all features");

// This reuses the FIRST cache entry (same URL + prompt)
webFetch.webFetch("https://example.com", "What is the main topic?");
```

--------------------------------

### Initial Conversation

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/memory/memory-filesystem-tools-demo/README.md

Start a conversation with the agent, providing initial information about yourself.

```text
USER: My name is Alice. I'm a backend engineer and I prefer short answers.
USER: We're migrating from PostgreSQL to CockroachDB this quarter.
USER: exit
```

--------------------------------

### Basic SkillsTool Builder Initialization

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

Demonstrates the most basic configuration of the SkillsTool using its builder pattern, specifying a single directory for skill loading.

```java
SkillsTool.builder()
    .addSkillsDirectory(".claude/skills")
    .build();
```

--------------------------------

### AgentEnvironment.info() Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/AgentEnvironment.md

Collects and prints current environment information, including working directory, git status, platform, OS version, and date. Ensure AgentEnvironment is imported.

```java
String envInfo = AgentEnvironment.info();
System.out.println(envInfo);
// Working directory: /Users/username/projects/myapp
// Is directory a git repo: Yes
// Platform: mac os x
// OS Version: Mac OS X 14.5.0
// Today's date: 2026-01-10
```

--------------------------------

### Configure AskUserQuestionTool with Custom Handler

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/ask-user-question-demo/README.md

Example of configuring the AskUserQuestionTool with a custom question handler in Application.java.

```java
tool = AskUserQuestionTool.builder()
    .questionHandler(new ConsoleQuestionHandler())
    .build();
```

--------------------------------

### Custom GrepTool Configuration with Limits

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GrepTool.md

Shows how to instantiate GrepTool with custom limits for output length, directory depth, and line length. Useful for large codebases or specific processing needs.

```java
GrepTool customGrepTool = new GrepTool(
    200000,  // maxOutputLength - Maximum output before truncation (default: 100000)
    50,      // maxDepth - Directory traversal depth (default: 100)
    5000     // maxLineLength - Max line length to process (default: 10000)
);
```

--------------------------------

### Find All Test Files

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Example of finding all files ending with 'Test.java' within the 'src/test' directory recursively.

```java
String testFiles = globTool.glob("**/*Test.java", "./src/test");
```

--------------------------------

### Configure ChatClient with FileSystemTools

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/memory/memory-filesystem-tools-demo/README.md

Sets up the ChatClient with default system prompt, memory directory, and includes ShellTools and FileSystemTools. Also configures ToolCallAdvisor and a custom MyLoggingAdvisor.

```java
ChatClient chatClient = chatClientBuilder
    .defaultSystem(p -> p
        .text(mainPrompt + "\n\n" + memoryToolsPrompt)
        .param("MEMORIES_ROOT_DIERCTORY", memoryDir))   // tells the agent where to write
    .defaultTools(
        ShellTools.builder().build(),          // Bash — for ls, mkdir, etc.
        FileSystemTools.builder().build())     // Read, Write, Edit — memory file operations
    .defaultAdvisors(
        ToolCallAdvisor.builder().build(),
        MyLoggingAdvisor.builder().build())
    .build();
```

--------------------------------

### Integrate GlobTool with FileSystemTools

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Demonstrates how to use GlobTool to find configuration files and then use FileSystemTools to read their content.

```java
GlobTool globTool = GlobTool.builder().build();
FileSystemTools fileTools = FileSystemTools.builder().build();

// Find all config files
String configs = globTool.glob("**/*.yml", "./config");

// Read each config
for (String configPath : configs.split("\n")) {
    String content = fileTools.read(configPath, null, null);
    // Process config content
}
```

--------------------------------

### Configure Agent with Tools and Advisors

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/index.md

Configure a Spring Boot application to build a ChatClient with various tools, sub-agents, and advisors. This example sets up task tools, skills, core utilities, and memory advisors.

```java
@SpringBootApplication
public class Application {

    @Bean
    CommandLineRunner demo(ChatClient.Builder chatClientBuilder,
        @Value("${BRAVE_API_KEY}") String braveApiKey,
        @Value("${agent.skills.paths}") List<Resource> skillPaths,
        @Value("classpath:/prompt/MAIN_AGENT_SYSTEM_PROMPT_V2.md") Resource agentSystemPrompt) {

        return args -> {
            // Configure Task tool with Claude sub-agents
            var taskTool = TaskTool.builder()
                .subagentTypes(ClaudeSubagentType.builder()
                    .chatClientBuilder("default", chatClientBuilder.clone())
                    .skillsResources(skillPaths)
                    .braveApiKey(braveApiKey)
                    .build())
                .build();

            ChatClient chatClient = chatClientBuilder
                // Main agent prompt
                .defaultSystem(p -> p.text(agentSystemPrompt)
                    .param(AgentEnvironment.ENVIRONMENT_INFO_KEY, AgentEnvironment.info())
                    .param(AgentEnvironment.GIT_STATUS_KEY, AgentEnvironment.gitStatus())
                    .param(AgentEnvironment.AGENT_MODEL_KEY, "claude-sonnet-4-5-20250929")
                    .param(AgentEnvironment.AGENT_MODEL_KNOWLEDGE_CUTOFF_KEY, "2025-01-01"))

                .defaultTools(
                    // Sub-Agents
                    taskTool,

                    // Skills
                    SkillsTool.builder().addSkillsResources(skillPaths).build(),

                    // Core Tools
                    ShellTools.builder().build(),
                    FileSystemTools.builder().build(),
                    GrepTool.builder().build(),
                    GlobTool.builder().build(),
                    SmartWebFetchTool.builder(chatClientBuilder.clone().build()).build(),
                    BraveWebSearchTool.builder(braveApiKey).build(),

                    // Task orchestration
                    TodoWriteTool.builder().build(),

                    // User feedback tool (use CommandLineQuestionHandler for CLI apps)
                    AskUserQuestionTool.builder()
                        .questionHandler(new CommandLineQuestionHandler())
                        .build())

                // Advisors
                .defaultAdvisors(
                    MessageChatMemoryAdvisor.builder(MessageWindowChatMemory.builder().maxMessages(500).build()).build())

                .build();

            String response = chatClient
                .prompt("Search for Spring AI documentation and summarize it")
                .advisors(a -> a.param(ChatMemory.CONVERSATION_ID, "session-1"))
                .call()
                .content();
        };
    }
}
```

--------------------------------

### Agent Configuration for Provider Routing

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/TaskTools.md

Example of configuring an agent to route to a different model provider, such as Ollama with Llama3.

```markdown
---
name: local-searcher
model: ollama:llama3
---

```

--------------------------------

### Use Relative Path with GlobTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Use a relative path to search for files starting from the current working directory.

```java
// Relative to current working directory
globTool.glob("*.java", "./src");
```

--------------------------------

### Agent Configuration with Disallowed Tools

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/docs/TaskTools.md

Example of configuring an agent with specific tools and explicitly disallowing dangerous ones for safety.

```markdown
---
name: explorer
tools: Read, Grep, Glob
disallowedTools: Edit, Write, Bash, Shell
---

```

--------------------------------

### Injecting and Using Environment Parameters

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/AgentEnvironment.md

Demonstrates how to configure agent properties, inject them using `@Value`, and provide them as parameters to the system prompt using AgentEnvironment.

```APIDOC
## Basic Usage

### 1. Configure Agent Properties

Add agent configuration to your `application.properties`:

```properties
# AGENT CONFIGURATION

## Model info (Must match the configured model above)
agent.model=claude-sonnet-4-5-20250929
agent.model.knowledge.cutoff=2025-01

# For GPT models
# agent.model=gpt-5-mini-2025-08-07
# agent.model.knowledge.cutoff=2025-08-07

# For Gemini models
# agent.model=gemini-3.1-pro-preview
# agent.model.knowledge.cutoff=Unknown
```

### 2. Inject Configuration Values

Use `@Value` annotations to inject the configuration:

```java
@Value("${agent.model:Unknown}")
String agentModel;

@Value("${agent.model.knowledge.cutoff:Unknown}")
String agentModelKnowledgeCutoff;
```

### 3. Configure System Prompt with Parameters

Use the `AgentEnvironment` utility to provide dynamic context to your system prompt:

```java
import org.springaicommunity.agent.utils.AgentEnvironment;

ChatClient chatClient = chatClientBuilder
    .defaultSystem(p -> p.text(systemPrompt) // Load system prompt from classpath
        .param(AgentEnvironment.ENVIRONMENT_INFO_KEY, AgentEnvironment.info())
        .param(AgentEnvironment.GIT_STATUS_KEY, AgentEnvironment.gitStatus())
        .param(AgentEnvironment.AGENT_MODEL_KEY, agentModel)
        .param(AgentEnvironment.AGENT_MODEL_KNOWLEDGE_CUTOFF_KEY, agentModelKnowledgeCutoff))
    // ... rest of configuration
    .build();
```

### 4. Reference Parameters in System Prompt

Create a system prompt template that references the parameters:

**File:** `src/main/resources/prompt/MAIN_AGENT_SYSTEM_PROMPT_V2.md`

```markdown
Here is useful information about the environment you are running in:
<env>
{ENVIRONMENT_INFO}
</env>
You are powered by the model: {AGENT_MODEL}

Assistant knowledge cutoff is {AGENT_MODEL_KNOWLEDGE_CUTOFF}.

{GIT_STATUS}
```
```

--------------------------------

### Multi-Pattern Search for Service and Repository Classes

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GrepTool.md

This example demonstrates how to find all service and repository classes using GrepTool by performing separate searches and combining the results. It utilizes `OutputMode.files_with_matches` and filters by Java files.

```java
GrepTool grepTool = GrepTool.builder().build();

// Find all service classes
String services = grepTool.grep(
    "@Service|@Component",
    "./src",
    null,
    OutputMode.files_with_matches,
    null, null, null, null, null, "java", null, null, null
);

// Find all repositories
String repositories = grepTool.grep(
    "@Repository",
    "./src",
    null,
    OutputMode.files_with_matches,
    null, null, null, null, null, "java", null, null, null
);

// Combine results for architectural analysis
```

--------------------------------

### Explicit Subagent Invocation

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/TaskTools.md

Users can explicitly request specific sub-agents by mentioning them in the prompt. This example invokes the code-reviewer subagent.

```java
String response = chatClient
    .prompt("Use the code-reviewer subagent to review my recent changes")
    .call()
    .content();
```

--------------------------------

### Find Dockerfile with GlobTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Use GlobTool to find 'Dockerfile' files, used for Docker containerization, starting from the current directory.

```java
// Docker
globTool.glob("**/Dockerfile", ".");
```

--------------------------------

### Set API Key and Run Application

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/subagent-a2a-demo/README.md

Set the Google GenAI API key as an environment variable and run the Spring Boot application using Maven.

```bash
export GOOGLE_GENAI_API_KEY=your-key
mvn spring-boot:run
```

--------------------------------

### Find All Markdown Files

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Use a simple glob pattern to find all files with the .md extension starting from the current directory.

```java
String mdFiles = globTool.glob("*.md", ".");
```

--------------------------------

### SubagentType Record

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/Subagent.md

Defines a subagent with a resolver and executor. It provides a convenience method to get the subagent's kind from its executor.

```java
public record SubagentType(
    SubagentResolver resolver,
    SubagentExecutor executor
) {
    public String kind() { return executor.getKind(); }
}
```

--------------------------------

### Initialize AskUserQuestionTool with CommandLineQuestionHandler

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/spring-ai-agent-utils/README.md

Shows how to initialize AskUserQuestionTool for CLI applications using CommandLineQuestionHandler.

```java
// For CLI applications, use the provided CommandLineQuestionHandler
AskUserQuestionTool askTool = AskUserQuestionTool.builder()
    .questionHandler(new CommandLineQuestionHandler())
    .build();
```

--------------------------------

### Use Simple Patterns

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GlobTool.md

Best practice for GlobTool: use simple patterns like '*.java' when possible. Explicitly using '**/*.java' is often redundant for simple file extensions and less clear.

```java
// ✅ GOOD: Simple pattern is faster
globTool.glob("*.java", "./src");

// ⚠️ UNNECESSARY: Explicit ** is redundant for simple extensions
globTool.glob("**/*.java", "./src");
// Both work the same, but simple pattern is clearer
```

--------------------------------

### Initialize BraveWebSearchTool

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/BraveWebSearchTool.md

Basic initialization of the BraveWebSearchTool with an API key. Use this to create an instance for performing web searches.

```java
BraveWebSearchTool searchTool = BraveWebSearchTool.builder("your-api-key")
    .build();

String results = searchTool.webSearch("Spring AI framework", null, null);
System.out.println(results);
```

--------------------------------

### TodoWrite Tool Result Example

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/examples/todo-demo/README.md

Displays the final output after the agent has completed all tasks, including the grouped and inverted movie titles.

```plaintext
Group 1
  pmuG tserroF (Forrest Gump)
  nayR etavirP gnivaS (Saving Private Ryan)

Group 2
  eliM neerG ehT (The Green Mile)
  yrotS yoT (Toy Story)

Group 3
  yawA tsaC (Cast Away)
  31 ollopA (Apollo 13)
...
```

--------------------------------

### Valid Skill Name Examples

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/SkillsTool.md

Skill names must be lowercase, alphanumeric, and use hyphens. They have a maximum length of 64 characters.

```yaml
# ✅ GOOD
name: pdf-processor
name: data-analysis
name: api-documentation

# ❌ BAD
name: PDF Processor      # No spaces or capitals
name: data_analysis      # Use hyphens, not underscores
name: my-super-long-skill-name-that-exceeds-the-limit  # Too long
```

--------------------------------

### GrepTool Basic Usage

Source: https://github.com/spring-ai-community/spring-ai-agent-utils/blob/main/docs/tools/GrepTool.md

Demonstrates the basic usage of the GrepTool with default configuration to search for a pattern in the current directory.

```APIDOC
## GrepTool Basic Usage

This example shows how to initialize the GrepTool with default settings and perform a basic search.

### Method
```java
String grep(String pattern, String path, String glob, OutputMode outputMode, Integer linesBefore, Integer linesAfter, Integer linesSurrounding, Boolean showLineNumbers, Boolean caseInsensitive, String type, Integer headLimit, Integer offset, Boolean multiline)
```

### Parameters
- **pattern** (String) - Required: Regex pattern to search.
- **path** (String) - Optional: Directory/file to search (default: current dir).
- **glob** (String) - Optional: Glob pattern to filter files (e.g., "*.java").
- **outputMode** (OutputMode) - Optional: Output format (default: files_with_matches).
- **linesBefore** (Integer) - Optional: Context lines before match (-B).
- **linesAfter** (Integer) - Optional: Context lines after match (-A).
- **linesSurrounding** (Integer) - Optional: Context lines before+after (-C).
- **showLineNumbers** (Boolean) - Optional: Show line numbers in content mode (-n).
- **caseInsensitive** (Boolean) - Optional: Case-insensitive search (-i).
- **type** (String) - Optional: File type filter (e.g., "java", "js").
- **headLimit** (Integer) - Optional: Limit output lines/entries.
- **offset** (Integer) - Optional: Skip first N lines/entries.
- **multiline** (Boolean) - Optional: Enable multiline matching.

### Request Example
```java
// Default configuration
GrepTool grepTool = GrepTool.builder().build();

// Search for pattern in current directory
String result = grepTool.grep(
    "TODO",              // pattern
    null,                // path (uses current directory)
    null,                // glob
    OutputMode.files_with_matches,  // outputMode
    null, null, null,    // context lines (before, after, surrounding)
    null,                // showLineNumbers
    null,                // caseInsensitive
    null,                // type
    null, null, null     // headLimit, offset, multiline
);
```
```