### Run vibe-tools Setup

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Executes the vibe-tools setup command to configure API keys, install Playwright browsers, and create/update AI instruction files for supported IDEs.

```bash
vibe-tools install .
```

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

### Install Vibe Tools Globally

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Install vibe-tools globally using npm. Run interactive setup for API keys and IDE integration, or use non-interactive setup for CI/CD.

```bash
# Install globally via npm
npm install -g vibe-tools

# Run interactive setup (configures API keys and IDE integration)
vibe-tools install .

# Non-interactive installation for CI/CD environments
CI=true vibe-tools install .
```

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

### Common Command Options Example

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Illustrates how to combine various common options for a vibe-tools command. This example performs a security audit on a repository.

```bash
vibe-tools repo "Security audit" --provider anthropic --model claude-sonnet-4-20250514 --max-tokens 16000 --save-to=security-audit.md --debug
```

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

### Repository Context Examples

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Examples of using the `vibe-tools repo` command to analyze codebases. You can specify subdirectories, remote repositories, and reasoning effort.

```bash
# Architecture understanding
vibe-tools repo "Explain the overall architecture of this project"

# Find usage examples
vibe-tools repo "Show me examples of error handling in this codebase"

# Debugging help
vibe-tools repo "Why might the authentication be failing in the login flow?"

# Analyze specific subdirectory
vibe-tools repo "Explain the code structure" --subdir=src/components

# Analyze remote GitHub repository
vibe-tools repo "Explain the architecture" --from-github=username/repo-name

# Deep analysis with enhanced reasoning
vibe-tools repo "Analyze the security implications of our authentication implementation" --reasoning-effort high

# Include web documentation as context
vibe-tools repo "Help me implement useState in my component" --with-doc=https://react.dev/reference/react/useState or a local file path"
```

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

### Web Search Examples

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Examples of using the `vibe-tools web` command for various search queries. You can specify a provider and reasoning effort for advanced searches.

```bash
# Get information about new technologies
vibe-tools web "What are the key features of Bun.js?"

# Check API documentation
vibe-tools web "How to implement OAuth2 in Express.js?"

# Compare technologies
vibe-tools web "Compare Vite vs Webpack for modern web development"

# Use XAI Grok for web search with reasoning
vibe-tools web "What are the latest AI safety developments?" --provider xai --reasoning-effort high
```

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

### Direct Model Query Examples

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Examples of using the `vibe-tools ask` command for direct model queries. You can specify the provider, model, and reasoning effort for detailed responses.

```bash
# Basic question
vibe-tools ask "What is the capital of France?" --provider openai --model o3-mini

# Complex algorithm explanation with high reasoning effort
vibe-tools ask "Explain the quicksort algorithm and analyze its time complexity in different scenarios" --provider openai --model o3-mini --reasoning-effort high

# Comparative analysis with Claude model and enhanced reasoning
vibe-tools ask "Compare and contrast microservices vs monolithic architecture" --provider anthropic --model claude-sonnet-4-20250514 --reasoning-effort medium
```

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

### Install vibe-tools Globally

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Installs the vibe-tools package globally using npm. This makes the CLI available system-wide.

```bash
npm install -g vibe-tools
```

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

### Manual Browser Installation

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

If automatic browser installation fails, you can manually install the necessary Playwright browsers using the provided command.

```bash
npx playwright install chromium
```

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

### Basic Vibe Tools Command Execution

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Illustrates the fundamental structure for running any command within the vibe-tools CLI, followed by a specific example for the 'web' command.

```bash
vibe-tools <command> [options]
```

```bash
vibe-tools web "What's new in TypeScript 5.7?"
```

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

### Serve Test Files for Browser Commands

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Starts a local test server for browser command testing. Place your HTML test files in the `tests/commands/browser/` directory. This is a long-running process.

```bash
pnpm serve-test
```

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

### Get Repository Context with Gemini 2.0

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Leverage Google Gemini 2.0 models with large context windows to get codebase-aware assistance and explanations for your project.

```bash
# Get context-aware assistance
vibe-tools repo "Explain the authentication flow in this project, which files are involved?"
```

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

### Execute 'cursor-tools --help' Command

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/reports/main/test-asset-handling_report_2025-03-11T14-58-23-881Z.md

Attempts to execute the help command for 'cursor-tools'. This example demonstrates a command failure scenario, indicating issues with how the help flag is processed.

```json
{
  "command": "cursor-tools --help"
}
```

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

### Execute 'cd' Command (Disallowed)

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/reports/main/test-asset-handling_report_2025-03-11T14-58-23-881Z.md

This example shows an attempt to execute the 'cd' command, which is explicitly disallowed by the tool's security policy. The output details the permitted commands.

```json
{
  "command": "cd /Users/andy/repos/cursor-tools-worktree/main && npm test -- --filter=test-asset-handling"
}
```

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

### Start Chrome for Debugging

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/browser/browser-command.md

Manually start a Chrome instance in debug mode on a specific port. This is a prerequisite for using vibe-tools to connect to an existing browser instance.

```bash
open -a "Google Chrome" --args --remote-debugging-port=9222 --no-first-run --no-default-browser-check --user-data-dir="/tmp/chrome-remote-debugging"
```

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

### Browser Observe: Page Overview

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Get a general overview of interactive elements on a web page. Useful for initial exploration.

```bash
vibe-tools browser observe "What can I interact with on this page?" --url "https://example.com"
```

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

### Example repomix.config.json for Inclusion and Compression

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Use a repomix.config.json file for advanced configuration, such as specifying which files to include and enabling output compression. This configuration is automatically detected by repo, plan, and doc commands.

```json
{
  "include": ["src/**/*", "README.md", "package.json"],
  "output": {
    "compress": true
  }
}
```

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

### vibe-tools JSON Configuration Example

Source: https://github.com/eastlondoner/vibe-tools/blob/main/CONFIGURATION.md

A comprehensive JSON configuration file for vibe-tools, detailing settings for various services like Perplexity, Gemini, Plan, Repo, Doc, Browser, and Token Counting. This file allows customization of AI models, token limits, and specific command behaviors.

```json
{
  "perplexity": {
    "model": "sonar-pro", // Default model for web search
    "maxTokens": 32000 // Maximum tokens for responses
  },
  "gemini": {
    "model": "gemini-2.5-pro-preview", // Default model for repository analysis
    "maxTokens": 32000 // Maximum tokens for responses
  },
  "plan": {
    "fileProvider": "gemini", // Provider for file identification
    "thinkingProvider": "openai", // Provider for plan generation
    "fileMaxTokens": 32000, // Tokens for file identification
    "thinkingMaxTokens": 32000 // Tokens for plan generation
  },
  "repo": {
    "provider": "gemini", // Default provider for repo command
    "maxTokens": 32000 // Maximum tokens for responses
  },
  "doc": {
    "maxRepoSizeMB": 100, // Maximum repository size for remote docs
    "provider": "gemini", // Default provider for doc generation
    "maxTokens": 32000 // Maximum tokens for responses
  },
  "browser": {
    "defaultViewport": "1280x720", // Default browser window size
    "timeout": 30000, // Default timeout in milliseconds
    "stagehand": {
      "env": "LOCAL", // Stagehand environment
      "headless": true, // Run browser in headless mode
      "verbose": 1, // Logging verbosity (0-2)
      "debugDom": false, // Enable DOM debugging
      "enableCaching": false, // Enable response caching
      "model": "claude-sonnet-4-20250514", // Default Stagehand model
      "provider": "anthropic", // AI provider (anthropic or openai)
      "timeout": 30000 // Operation timeout
    }
  },
  "tokenCount": {
    "encoding": "o200k_base" // Token counting method
  },
  "openai": {
    "maxTokens": 32000 // Will be used when provider is "openai"
  },
  "anthropic": {
    "maxTokens": 21000 // Will be used when provider is "anthropic"
  },
  "groq": {
    "model": "llama-3.3-70b-versatile",
    "maxTokens": 16384
  }
}
```

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

### Flask RESTful API Setup

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/ask/groq-provider-tests/large-context-doc.txt

Initializes a Flask application with SQLAlchemy for database interaction and Flask-RESTful for API routing. Configure database URI and JWT secret key.

```python
from flask import Flask, request, jsonify
from flask_restful import Api, Resource
from flask_sqlalchemy import SQLAlchemy
from flask_jwt_extended import JWTManager, create_access_token, jwt_required, get_jwt_identity
from werkzeug.security import generate_password_hash, check_password_hash
from datetime import datetime, timedelta
import re

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'postgresql://user:password@localhost/api_db'
app.config['JWT_SECRET_KEY'] = 'your-secret-key'
app.config['JWT_ACCESS_TOKEN_EXPIRES'] = timedelta(hours=24)

db = SQLAlchemy(app)
api = Api(app)
jwt = JWTManager(app)
```

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

### Environment Variable Setup

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/web/xai-web-search-tests.md

Sets the necessary environment variable for XAI API authentication. Ensure this is configured before running XAI-related tests.

```bash
export XAI_API_KEY="your-xai-api-key"
```

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

### Example Feature Behavior File

Source: https://github.com/eastlondoner/vibe-tools/blob/main/docs/TESTING.md

Defines a test scenario for the web command in vibe-tools. This Markdown file specifies the task, expected behavior, and success criteria for the AI agent.

```markdown
# Feature Behavior: Web Command

## Description

This test verifies that the web command can query online information correctly.

## Test Scenarios

### Scenario 1: Basic Web Query

**Task Description:**
Use vibe-tools to search for information about climate change.

**Expected Behavior:**

- The AI agent should use the web command
- Response should include information about climate change
- The command should complete successfully

**Success Criteria:**

- AI agent correctly uses the vibe-tools web command
- Response contains relevant information about climate change
- Command completes without errors
```

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

### Describe Table Schema with MCP Run

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/mcp/mcp-command.md

Use vibe-tools to execute the describe_table tool with the SQLite MCP server to get schema information for a specific table.

```bash
vibe-tools mcp run --tool describe_table --table "users" --db "{{path:test.db}}" -- "ONLY mcp-server-sqlite should be used"
```

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

### Asset Storage Structure Example

Source: https://github.com/eastlondoner/vibe-tools/blob/main/WIP.md

Illustrates the directory structure for storing test assets relative to their corresponding test files. Assets are placed in a subdirectory named after the test file.

```tree
tests/feature-behaviors/
    ask/
        ask-command.md              # Test file
        ask-command/                # Assets directory for this test
            scenario9-long-query.txt
    browser/
        browser-open-command.md     # Test file
        browser-open-command/       # Assets directory for this test
            test-form.html
```

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

### Run MCP Server with OpenRouter

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Use OpenRouter as the provider for an MCP server task, specifying the model to be used. This example fetches weather data.

```bash
vibe-tools mcp run "api-client" "Fetch weather data" --provider openrouter --model openai/gpt-4o
```

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

### Python Microservice with Flask and SQLAlchemy

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/ask/groq-provider-tests/large-context-doc.txt

Example of a simple microservice using Flask for web requests and SQLAlchemy for database interactions. It defines a User model and provides endpoints for retrieving and creating users. Ensure DATABASE_URL environment variable is set for production.

```python
from flask import Flask, jsonify, request
from flask_sqlalchemy import SQLAlchemy
import os

app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = os.environ.get('DATABASE_URL', 'sqlite:///users.db')
db = SQLAlchemy(app)

class User(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    username = db.Column(db.String(80), unique=True, nullable=False)
    email = db.Column(db.String(120), unique=True, nullable=False)
    created_at = db.Column(db.DateTime, default=db.func.now())

    def to_dict(self):
        return {
            'id': self.id,
            'username': self.username,
            'email': self.email,
            'created_at': self.created_at.isoformat()
        }

@app.route('/users', methods=['GET'])
def get_users():
    users = User.query.all()
    return jsonify([user.to_dict() for user in users])

@app.route('/users', methods=['POST'])
def create_user():
    data = request.get_json()
    user = User(username=data['username'], email=data['email'])
    db.session.add(user)
    db.session.commit()
    return jsonify(user.to_dict()), 201

@app.route('/users/<int:user_id>', methods=['GET'])
def get_user(user_id):
    user = User.query.get_or_404(user_id)
    return jsonify(user.to_dict())

if __name__ == '__main__':
    with app.app_context():
        db.create_all()
    app.run(debug=True, port=5001)
```

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

### Open URL and Capture Content with vibe-tools browser

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

The 'browser open' command opens a URL and can capture HTML content, console logs, network activity, take screenshots, or start an interactive debugging session.

```bash
vibe-tools browser open "https://example.com" --html
```

```bash
vibe-tools browser open "https://example.com" --screenshot=page.png
```

```bash
vibe-tools browser open "https://example.com" --connect-to=9222
```

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

### Execute MCP Server Tools

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Executes MCP server tools using natural language queries. The query must provide enough context for vibe-tools to select the appropriate server. Results are streamed in real-time. Example: vibe-tools mcp run "list files in the current directory" --provider=openrouter

```bash
vibe-tools mcp run "<query>"
```

```bash
vibe-tools mcp run "using the mcp-server-sqlite list files in directory --provider=openrouter"
```

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

### Open URL in Browser

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Opens a specified URL in a browser instance and captures page content, console logs, and network activity. Use options like --html to capture specific data. Example: vibe-tools browser open "https://example.com" --html

```bash
vibe-tools browser open <url> [options]
```

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

### Create Implementation Plan from YouTube Tutorial

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Generate an implementation plan based on the content of a coding tutorial video. This helps in structuring the steps required to follow the tutorial.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=dQw4w9WgXcQ" --type=plan
```

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

### Generate Basic Implementation Plan

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Create a basic implementation plan for a given task using the 'plan' command.

```bash
vibe-tools plan "Add user authentication to the login page"
```

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

### Analyze YouTube Video for Implementation Plan

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Use this command to generate an implementation plan from a YouTube video. Ensure the video URL is correct.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=VIDEO_ID" --type=plan
```

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

### Application Initialization and Run

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/ask/groq-provider-tests/large-context-doc.txt

Initializes the application context, creates database tables if they don't exist, and runs the Flask development server.

```python
if __name__ == '__main__':
    with app.app_context():
        db.create_all()
    app.run(debug=True)
```

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

### Repomix Ignore File Example for Laravel

Source: https://github.com/eastlondoner/vibe-tools/blob/main/CONFIGURATION.md

Example .repomixignore file for a Laravel project to exclude specific directories and files from documentation analysis.

```ignore
vendor/
public/
database/
storage/
.idea
.env
```

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

### Plan with Multiple Documentation Sources

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Provide multiple documentation sources, both local and remote, to enhance the implementation plan.

```bash
vibe-tools plan "Build REST API" --with-doc=./api-design.md --with-doc=https://restfulapi.net/
```

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

### Non-Interactive vibe-tools Installation for CI/CD

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Installs vibe-tools in a non-interactive mode, suitable for CI/CD environments. It automatically detects package managers and IDEs, uses existing configurations, and relies on environment variables for API keys.

```bash
CI=true vibe-tools install .
```

```bash
NONINTERACTIVE=true vibe-tools install .
```

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

### User Resource - GET

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/ask/groq-provider-tests/large-context-doc.txt

Retrieves a specific user by ID. Requires JWT authentication.

```python
class UserResource(Resource):
    @jwt_required()
    def get(self, user_id):
        user = User.query.get_or_404(user_id)
        return user.to_dict(), 200
```

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

### Plan with External Documentation Context

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Incorporate external documentation from a URL to inform the implementation plan generation.

```bash
vibe-tools plan "Integrate Stripe payments" --with-doc=https://stripe.com/docs/payments/quickstart
```

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

### YouTube Analysis: Get Transcript

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Retrieve the detailed transcript of a YouTube video. Specify the type as 'transcript'.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=VIDEO_ID" --type=transcript
```

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

### Basic Groq Provider Query

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/ask/groq-provider-tests.md

Use the vibe-tools ask command with the groq provider and a specific model. Ensure the command executes successfully and returns a coherent response.

```bash
vibe-tools ask "Explain the concept of recursion in programming with a simple example" --provider groq --model moonshotai/kimi-k2-instruct
```

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

### Linear Integration: Get Issue

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Retrieve a specific Linear issue using its identifier (e.g., ITE-123) or UUID.

```bash
vibe-tools linear get-issue ITE-123
```

```bash
vibe-tools linear issue ABC-456
```

```bash
vibe-tools linear get-issue 0b906a8e-f8a8-477d-9c98-0986b09ac5f9
```

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

### Global Rule Configuration

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Example configuration for a global rule that should always be applied. This is often used for foundational settings or policies.

```yaml
---
description: Global Rule. This rule should ALWAYS be loaded
globs: *,**/*
alwaysApply: true
---
```

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

### Ask Specific Questions About a YouTube Video

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Query a YouTube video for specific information. Replace 'What authentication method does the tutorial use?' with your question.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=VIDEO_ID" "What authentication method does the tutorial use?"
```

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

### Observe Webpage Elements

Source: https://github.com/eastlondoner/vibe-tools/blob/main/CLAUDE.md

Observe interactive elements on a webpage and get suggestions for possible actions. Specify the URL to observe.

```bash
vibe-tools browser observe "<instruction>" --url=<url> [options]
```

```bash
vibe-tools browser observe "interactive elements" --url=https://example.com
```

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

### Handle Login Button Click

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/commands/browser/interactive-test.html

Logs 'Login clicked' to the console when the login button is pressed. No additional setup is required.

```javascript
function handleLogin() { console.log('Login clicked'); }
```

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

### Ask Groq's moonshotai/kimi-k2-instruct Model

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Query Groq's specific model for explanations. This command is suitable for general knowledge queries.

```bash
vibe-tools ask "Explain quantum computing" --provider groq --model moonshotai/kimi-k2-instruct
```

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

### Get GitHub PR Metadata Only

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Retrieve only the metadata (labels, assignees, etc.) for a GitHub pull request. Replace '123' with the actual PR number.

```bash
vibe-tools github pr 123 --metadata-only
```

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

### Plan with Enhanced Reasoning

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Increase the reasoning effort for generating a more thorough implementation plan.

```bash
vibe-tools plan "Refactor authentication system" --reasoning-effort high
```

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

### User List Resource - GET

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/ask/groq-provider-tests/large-context-doc.txt

Retrieves a paginated list of active users. Supports 'page' and 'per_page' query parameters. Requires JWT authentication.

```python
class UserListResource(Resource):
    @jwt_required()
    def get(self):
        page = request.args.get('page', 1, type=int)
        per_page = request.args.get('per_page', 10, type=int)
        
        users = User.query.filter_by(is_active=True).paginate(
            page=page, per_page=per_page, error_out=False
        )
        
        return {
            'users': [user.to_dict() for user in users.items],
            'total': users.total,
            'pages': users.pages,
            'current_page': page,
            'per_page': per_page
        }, 200
```

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

### Implement Feature with Diff and Doc

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Use the 'repo' command to implement a feature based on specifications, including generating diffs and incorporating documentation.

```bash
vibe-tools repo "Implement the feature per spec" --with-diff --with-doc=./feature-spec.md --base=develop
```

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

### Log Different Message Types

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/commands/browser/console-log.html

Use console.log, console.info, console.warn, and console.error to log messages with different severity levels. No special setup is required.

```javascript
console.log('Basic log message');
console.info('Info message');
console.warn('Warning message');
console.error('Error message');
```

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

### Perform Custom Analysis on YouTube Video

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Execute custom analysis on a YouTube video, such as extracting and explaining code examples. Specify your custom analysis prompt.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=VIDEO_ID" --type=custom "Extract all code examples and explain them"
```

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

### Execute Vibe-Tools with a File Input

Source: https://github.com/eastlondoner/vibe-tools/blob/main/TESTING.md

This command demonstrates how to use the vibe-tools CLI to process a query from a specified file. Ensure the file path is absolute to avoid issues.

```bash
vibe-tools ask --provider openai --model gpt-4 --file=/absolute/path/to/scenario9-long-query.txt
```

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

### Include Web Documentation as Context for Repo Analysis

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Analyze a repository's content while including external web documentation as context. This is useful for ensuring implementations align with specifications.

```bash
vibe-tools repo "Help me implement useState in my component" --with-doc=./path/to/local/docs.md
```

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

### Ask Question with Groq Provider

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Query an AI model using the Groq provider for fast inference.

```bash
vibe-tools ask "Explain quantum computing basics" --provider groq --model llama-3.3-70b-versatile
```

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

### Generate Critical Review of YouTube Tutorial

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Create a critical review of a tutorial's accuracy and quality. This is useful for assessing the educational value of a video.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=dQw4w9WgXcQ" --type=review
```

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

### Set CI/CD environment variables for vibe-tools

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Use these environment variables to ensure vibe-tools installs correctly in automated environments. Set `CI=true` or `NONINTERACTIVE=true` to prevent interactive prompts.

```bash
CI=true vibe-tools install .
# or
NONINTERACTIVE=true vibe-tools install .
```

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

### Generate Documentation for Current Repository

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Generate documentation for the current local repository and save it to a specified file.

```bash
vibe-tools doc --save-to=docs/README.md
```

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

### Wait Command in Scripting Context

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Demonstrates using the wait command in a script to introduce a delay between browser actions. This example clicks submit, waits, then opens the result page.

```bash
vibe-tools browser act "Click Submit" --url "https://example.com" && vibe-tools wait 3 && vibe-tools browser open "https://example.com/result" --screenshot=result.png
```

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

### Search for MCP Servers

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Search the marketplace for MCP (Model Context Protocol) servers. Use keywords like 'database' or 'file system' to find relevant servers.

```bash
vibe-tools mcp search "database"
```

```bash
vibe-tools mcp search "file system"
```

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

### Linear Integration: Connect

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Set up authentication for the Linear integration. This is typically an interactive process.

```bash
vibe-tools linear connect
```

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

### Ask Basic Question with Default Provider

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Pose a general question to an AI model using the default provider and a specified model.

```bash
vibe-tools ask "What is dependency injection?" --provider openai --model gpt-4o
```

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

### Get Specific Linear Issue

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Retrieve and display a specific Linear issue with all its details, including title, description, status, comments, and attachments. Supports alternative command names.

```bash
vibe-tools linear get-issue ITE-123
```

```bash
vibe-tools linear issue ABC-456
```

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

### Verify Asset Handling with Shell Commands

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/test/test-asset-handling.md

This snippet demonstrates a sequence of shell commands used to verify asset handling. It includes checking the current directory, displaying file content, listing files in a directory, and searching for specific file types.

```shell
pwd
```

```shell
cat {{path:sample-asset.txt}}
```

```shell
ls -la assets
```

```shell
find . -name "*.txt"
```

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

### Kill Process on Busy Port

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Use this command to kill any process listening on port 3000, typically used when `pnpm serve-test` fails to start due to a busy port.

```bash
lsof -i :3000 | grep LISTEN | awk '{print $2}' | xargs kill
```

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

### Configure Plan Command Models

Source: https://github.com/eastlondoner/vibe-tools/blob/main/CONFIGURATION.md

Configure the file identification and thinking models, their providers, and token limits for the plan command. Ensure the specified providers and models are available.

```json
{
  "plan": {
    "fileProvider": "gemini",
    "thinkingProvider": "openai",
    "fileModel": "gemini-2.5-pro-preview",
    "thinkingModel": "o3",
    "fileMaxTokens": 8192,
    "thinkingMaxTokens": 8192
  }
}
```

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

### Select AI Model for Browser Command

Source: https://github.com/eastlondoner/vibe-tools/blob/main/CONFIGURATION.md

Use the --model option to specify the AI model for the vibe-tools browser command. Examples show gpt-4o and Claude 3.7 Sonnet.

```bash
# Use gpt-4o
vibe-tools browser act "Click Login" --url "https://example.com" --model=gpt-4o
```

```bash
# Use Claude 3.7 Sonnet
vibe-tools browser act "Click Login" --url "https://example.com" --model=claude-sonnet-4-20250514
```

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

### Analyze Interactive Elements with vibe-tools browser observe

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Use 'browser observe' to get an overview of interactive elements on a page or to find specific elements using natural language queries.

```bash
vibe-tools browser observe "What can I interact with?" --url "https://example.com"
```

```bash
vibe-tools browser observe "Find the login form" --url "https://example.com"
```

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

### Browser Observe: Buttons and Actions

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

List all buttons on a page and describe their purposes.

```bash
vibe-tools browser observe "List all buttons and their purposes" --url "https://app.example.com"
```

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

### Get ClickUp Task Information

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Retrieves detailed information about a specific ClickUp task, including its description, comments, status, assignees, and metadata. Ensure the task ID is correctly formatted.

```bash
vibe-tools clickup task <task_id>
```

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

### Browser Observe: Navigation Options

Source: https://context7.com/eastlondoner/vibe-tools/llms.txt

Analyze a page to identify available navigation options.

```bash
vibe-tools browser observe "What navigation options are available?" --url "https://example.com"
```

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

### GitHub Actions Workflow for Continuous Testing

Source: https://github.com/eastlondoner/vibe-tools/blob/main/docs/TESTING.md

This YAML workflow automates the installation of dependencies and execution of tests on every push event. It ensures that vibe-tools commands function correctly as code changes are made.

```yaml
# GitHub Actions workflow
name: Tests
on: [push]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2
      - uses: actions/setup-node@v2
        with:
          node-version: '18'
      - run: npm install -g pnpm
      - run: pnpm install
      - run: pnpm dev test "tests/feature-behaviors/**/*.md" --parallel 4
```

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

### Get YouTube Video Transcript with Save Option

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Retrieve a full transcript with speaker annotations from a YouTube video and save it to a markdown file. Specify the output file name using --save-to.

```bash
vibe-tools youtube "https://www.youtube.com/watch?v=dQw4w9WgXcQ" --type=transcript --save-to=transcript.md
```

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

### Generate Documentation

Source: https://github.com/eastlondoner/vibe-tools/blob/main/AGENTS.md

Generates documentation for local or remote repositories.

```bash
vibe-tools doc
```

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

### Connect to Existing Browser and Take Screenshot

Source: https://github.com/eastlondoner/vibe-tools/blob/main/tests/feature-behaviors/browser/browser-command.md

Connect to a running Chrome instance on port 9222, navigate to a URL, and capture a screenshot. Ensure Chrome is started with the debug port enabled before running this command.

```bash
vibe-tools --connect-to http://localhost:9222 --url https://example.com --screenshot example.png
```

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

### Document Specific Aspects and Save to File

Source: https://github.com/eastlondoner/vibe-tools/blob/main/README.md

Generate documentation for specific aspects of a project and save it to a file without displaying output to stdout. Use the --hint flag to guide the documentation focus.

```bash
vibe-tools doc --save-to=docs/api.md --quiet --hint="Focus on the API endpoints and their usage"
```

```bash
vibe-tools doc --save-to=docs/architecture.md --quiet --hint="Focus on system architecture"
```