### Install, Build, and Run ACE MCP Server

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/START_HERE.md

This snippet shows the necessary bash commands to install project dependencies, build the TypeScript project, and start the ACE MCP server. Ensure you are in the project directory before running these commands.

```bash
# 1. Go to directory
cd $HOME/code/perplexity/ace-mcp-server

# 2. Install dependencies
npm install

# 3. Build project
npm run build

# 4. Test
npm start
```

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

### Clone and Setup ACE MCP

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/README.md

Instructions to clone the ACE MCP repository and set up initial configuration files. This involves cloning the Git repository, navigating into the directory, and copying an example environment file.

```bash
git clone https://github.com/Angry-Robot-Deals/ace-mcp.git
cd ace-mcp
cp .env.example .env
# Edit .env with your configuration
```

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

### Clone Repository and Setup Dev Environment (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/CONTRIBUTING.md

Commands to clone the ACE MCP Server repository locally and set up the development environment using Docker Compose. This involves copying an example environment file, editing it with specific configurations, and starting the development services.

```bash
git clone https://github.com/YOUR_USERNAME/ace-mcp.git
cd ace-mcp
cp .env.example .env
# Edit .env with your configuration
docker-compose -f docker-compose.dev.yml up -d
```

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

### Run ACE MCP Server

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INSTALLATION.md

Starts the ACE MCP Server. Can be run directly using 'npm start' for production or 'npm run dev' for development mode.

```bash
# Test standalone
npm start

# Or in development mode
npm run dev
```

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

### Install Dependencies and Build Project

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DESCRIPTION.md

These commands are essential for setting up the project environment. 'npm install' downloads and installs all the necessary Node.js packages, while 'npm run build' compiles the source code, typically TypeScript, into executable JavaScript.

```bash
cd $HOME/code/perplexity/ace-mcp-server
npm install
npm run build
```

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

### Verify Dependencies Installed (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/ASSETS_CHECKLIST.md

Confirms that the '@modelcontextprotocol/sdk' package has been successfully installed within the 'node_modules' directory. This is crucial for the project to function correctly.

```bash
ls node_modules/@modelcontextprotocol/sdk
# Should exist
```

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

### Manual Deployment: Start Services

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Starts the ACE MCP Server services using Docker Compose after the application has been extracted to the `/opt/ace-mcp-server` directory. This command runs the services in detached mode.

```bash
cd /opt/ace-mcp-server
docker-compose up -d
```

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

### Manual Deployment: Install Server Dependencies

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Installs essential dependencies on the production server, including Docker, Docker Compose, and Nginx. This is a prerequisite for running the ACE MCP Server in a manual deployment scenario.

```bash
# Install Docker
curl -fsSL https://get.docker.com -o get-docker.sh
sh get-docker.sh

# Install Docker Compose
curl -L "https://github.com/docker/compose/releases/latest/download/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
chmod +x /usr/local/bin/docker-compose

# Install Nginx
apt-get update
apt-get install -y nginx
```

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

### Start Server with npm

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DESCRIPTION.md

This command initiates the server process after it has been built. It relies on the package.json scripts to start the application.

```bash
npm start
```

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

### Automated Production Deployment Script

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Executes an automated deployment script (`deploy.sh`) to handle the entire production deployment process. This includes verifying environment, building Docker images, creating a deployment package, uploading to the server, installing dependencies, configuring Nginx, and starting services.

```bash
# Make script executable
chmod +x deploy.sh

# Run deployment
./deploy.sh
```

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

### Troubleshoot Services Won't Start

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Commands to diagnose and resolve issues where Ace MCP services fail to start. This includes checking Docker logs, verifying container status, and rebuilding Docker images.

```bash
# Check logs
docker-compose logs

# Check Docker status
docker ps -a

# Rebuild images
docker-compose up -d --build
```

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

### Install Project Dependencies (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INITIALIZATION_REPORT.md

Installs necessary project dependencies for the ace-mcp-server, including production and development packages. It navigates to the project directory and uses npm to manage packages.

```bash
cd $HOME/code/perplexity/ace-mcp-server
npm install openai axios zod
npm install -D @types/axios jest ts-jest @types/jest
```

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

### Start ACE MCP Server with Docker Compose

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_SETUP.md

This command initiates the ACE MCP server using Docker Compose in development mode. It requires the user to navigate to the server's directory first. Ensure Docker is installed and running.

```bash
cd $HOME/code/perplexity/ace-mcp-server
docker-compose -f docker-compose.dev.yml up -d
```

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

### Example Usage: Contextual Assistance (Command Prompt)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_AUTO_SETUP.md

Shows how to leverage the `@ace_context_aware` command for contextual assistance in Cursor AI. The example demonstrates a query related to API optimization and the expected output from ACE.

```plaintext
```
@ace_context_aware how to optimize API for high load domain:api
```
```

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

### Install Project Dependencies

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INSTALLATION.md

Installs project dependencies including @modelcontextprotocol/sdk, fs-extra, uuid, and TypeScript. This command should be run after navigating to the project directory.

```bash
cd $HOME/code/perplexity/ace-mcp-server
npm install
```

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

### View Server Resource Usage with htop

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Connects to the production server via SSH and launches 'htop' to monitor overall server resource utilization.

```bash
ssh -i ~/.ssh/id_ed25519 root@$PRODUCTION_SERVER_IP \
  'htop'
```

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

### Configure LLM Providers using Factory (TypeScript)

Source: https://context7.com/angry-robot-deals/ace-mcp/llms.txt

This snippet demonstrates how to create and configure various Large Language Model (LLM) providers using the `createLLMProvider` factory function. It shows examples for DeepSeek, OpenAI, Anthropic Claude, Google Gemini, Mistral, and LM Studio (local). It also includes examples of how to use the configured providers for chat and embedding generation.

```typescript
import { createLLMProvider } from './src/llm/factory';

// DeepSeek provider (recommended)
const deepseek = createLLMProvider({
  provider: 'deepseek',
  deepseek: {
    apiKey: 'sk-xxx',
    model: 'deepseek-chat',
    timeout: 30000,
    maxRetries: 3
  }
});

// OpenAI provider
const openai = createLLMProvider({
  provider: 'openai',
  openai: {
    apiKey: 'sk-xxx',
    model: 'gpt-4o',
    embeddingModel: 'text-embedding-3-small',
    timeout: 30000
  }
});

// Anthropic Claude provider
const claude = createLLMProvider({
  provider: 'anthropic',
  anthropic: {
    apiKey: 'sk-ant-xxx',
    model: 'claude-3-5-sonnet-20241022',
    maxTokens: 4096
  }
});

// Google Gemini provider
const gemini = createLLMProvider({
  provider: 'gemini',
  gemini: {
    apiKey: 'your-api-key',
    model: 'gemini-1.5-pro'
  }
});

// Mistral provider
const mistral = createLLMProvider({
  provider: 'mistral',
  mistral: {
    apiKey: 'your-api-key',
    model: 'mistral-large-latest'
  }
});

// LM Studio (local)
const lmstudio = createLLMProvider({
  provider: 'lmstudio',
  lmstudio: {
    baseUrl: 'http://localhost:1234/v1',
    model: 'local-model'
  }
});

// Use provider
const response = await deepseek.chat([
  { role: 'system', content: 'You are a helpful coding assistant.' },
  { role: 'user', content: 'Explain async/await in JavaScript' }
], {
  temperature: 0.7,
  maxTokens: 500
});

console.log('Response:', response);

// Generate embeddings
const embedding = await deepseek.embed('This is a sample text for embedding');
console.log('Embedding dimension:', embedding.length);
```

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

### Troubleshooting Build Failures

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INSTALLATION.md

Steps to resolve build failures, including checking file integrity, verifying Node.js version, and performing a clean install/build cycle.

```bash
# Verify Node.js version (should be >=18)
node --version

# Clean install and build
rm -rf node_modules dist
npm install
npm run build
```

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

### Install ACE MCP Server Dependencies

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_SETUP.md

These commands install necessary Node.js dependencies for the ACE MCP project, including the `tsx` runtime for TypeScript execution and project-specific packages. Run these in the project's root directory.

```bash
# In project directory
npm install -g tsx
npm install
```

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

### Install Project Dependencies (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/PROJECT_STATUS.md

This bash script outlines the commands to install project dependencies for the ACE MCP Server. It includes installing core packages like `openai`, `axios`, and `zod`, as well as development dependencies for testing such as `jest` and `ts-jest`.

```bash
# 1. Install dependencies
cd $HOME/code/perplexity/ace-mcp-server
npm install openai axios zod
npm install -D @types/axios jest ts-jest @types/jest

# 2. Create structure
mkdir -p src/llm src/llm/__tests__
mkdir -p src/core src/mcp src/storage src/utils

# 3. Start Phase 2
# Create src/llm/provider.ts (interface)
```

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

### Install NPM Dependencies for ACE MCP Server

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DESCRIPTION.md

This command installs all the necessary Node Package Manager (NPM) dependencies required to run the ACE MCP server. It reads the dependencies listed in the 'package.json' file and downloads them into the 'node_modules' directory.

```bash
npm install
```

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

### Start Enhanced MCP Server (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_AUTO_SETUP.md

These bash commands are used to stop the old MCP server and start the enhanced ACE MCP server. It involves navigating to the server directory and executing the enhanced server script.

```bash
cd $HOME/code/perplexity/ace-mcp-server

# Stop old server
pkill -f "mcp-server-simple"

# Start enhanced server
npx tsx src/mcp-server-enhanced.ts
```

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

### Example Usage: Prompt Enhancement (Command Prompt)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_AUTO_SETUP.md

Demonstrates the usage of the `@ace_enhance_prompt` command in Cursor AI for improving complex requests. The example shows how ACE adds security considerations and best practices to a prompt for creating a REST API.

```plaintext
```
@ace_enhance_prompt create REST API for users focus_area:security
```
```

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

### Build TypeScript Project

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INSTALLATION.md

Compiles TypeScript source files into JavaScript, outputting the compiled code to the 'dist/' directory. This is a crucial step before running the server.

```bash
npm run build
```

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

### Docker Compose Override for LLM Provider

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/LLM_PROVIDERS.md

An example of overriding service environment variables in a Docker Compose setup to configure the LLM provider and API key for the ACE server. It specifies the provider, API key, and model for DeepSeek.

```yaml
# docker-compose.override.yml
services:
  ace-server:
    environment:
      - LLM_PROVIDER=deepseek
      - DEEPSEEK_API_KEY=${DEEPSEEK_API_KEY}
      - DEEPSEEK_MODEL=deepseek-reasoner  # Use thinking mode
```

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

### Environment Variable Configuration Example (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INITIALIZATION_REPORT.md

Illustrates environment variable settings for LLM provider configuration. This approach aligns with 12-factor app methodology and is Docker-friendly, enabling type-safe validation through tools like Zod. It specifies the active LLM provider and its corresponding API keys or base URLs.

```bash
LLM_PROVIDER=openai|lmstudio
OPENAI_API_KEY=sk-...
LMSTUDIO_BASE_URL=http://10.242.247.136:11888/v1
```

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

### Initialize and Push GitHub Repository (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/GITHUB_INITIALIZATION.md

Commands executed to initialize a new Git repository, stage all files, create the initial commit, rename the branch to main, add a remote origin, and push the branch to GitHub. This assumes a specific SSH configuration for the remote.

```bash
git init
git add .
git commit -m "first commit: ..."
git branch -M main
git remote add origin git@github-ard:...
git push -u origin main
```

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

### Check .env File and Required Variables (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/deployment/DEPLOYMENT_README.md

This snippet demonstrates how to check the contents of the `.env` file to troubleshoot missing environment variables during deployment. It also lists the critical variables that must be set for the application to function correctly, such as API credentials and server details. Ensure all listed variables are present and correctly configured in your `.env` file.

```bash
# Check .env file
cat .env

# Verify required variables are set:
# - API_BEARER_TOKEN
# - PRODUCTION_SERVER_IP
# - PRODUCTION_DOMAIN
# - LLM_PROVIDER and its credentials
```

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

### Verify Configuration Valid (Node.js/Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/ASSETS_CHECKLIST.md

Executes the built JavaScript code to ensure it initializes without runtime errors. This verifies the project's configuration and basic functionality. Use Ctrl+C to exit.

```bash
node -e "require('./dist/index.js')"
# Should start without errors (Ctrl+C to exit)
```

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

### Environment Variables for Multi-Provider Setup

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/LLM_PROVIDERS.md

Configures primary and fallback LLM providers using environment variables. This setup allows specifying an API key for the main provider and another for fallback embeddings if the primary doesn't support them.

```env
# Primary provider for most tasks
LLM_PROVIDER=deepseek
DEEPSEEK_API_KEY=sk-xxx

# Fallback for embeddings (if primary doesn't support)
OPENAI_API_KEY=sk-xxx
OPENAI_EMBEDDING_MODEL=text-embedding-3-small
```

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

### Configure OpenAI LLM Provider

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/LLM_PROVIDERS.md

Environment variables for configuring the OpenAI LLM provider. This setup includes specifying the provider, API key, the chat model, and the embedding model.

```env
LLM_PROVIDER=openai
OPENAI_API_KEY=sk-your-api-key-here
OPENAI_MODEL=gpt-4o
OPENAI_EMBEDDING_MODEL=text-embedding-3-small
```

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

### npm Dependency Installation for LLM Features (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INITIALIZATION_REPORT.md

Lists the npm packages required to implement the LLM provider abstraction and related functionalities. This includes core libraries like 'openai', 'axios', and 'zod', as well as development dependencies for testing ('jest', 'ts-jest') and TypeScript support.

```bash
npm install openai axios zod
npm install -D @types/axios jest ts-jest @types/jest
```

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

### Docker Development Environment for ACE MCP

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/README.md

Commands to manage the Docker development environment for ACE MCP. This includes starting, viewing logs, and stopping the containerized development setup using Docker Compose.

```bash
# Start development environment
docker-compose -f docker-compose.dev.yml up -d

# View logs
docker-compose -f docker-compose.dev.yml logs -f

# Stop environment  
docker-compose -f docker-compose.dev.yml down
```

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

### Manual Deployment: Prepare Deployment Package

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Manually prepares the ACE MCP Server for deployment by building Docker images, creating a deployment directory, copying necessary files, and packaging them into a compressed archive (`.tar.gz`).

```bash
# Build Docker images
docker-compose build

# Create deployment directory
mkdir -p deploy-package
cp -r src dashboard package*.json tsconfig.json docker-compose.yml Dockerfile .env deploy-package/

# Create archive
tar -czf ace-mcp-deploy.tar.gz deploy-package/
```

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

### Configure LM Studio Local LLM Provider

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/LLM_PROVIDERS.md

Environment variables for configuring LM Studio as a local LLM provider. This setup uses a local base URL and specifies the model to be loaded by LM Studio.

```env
LLM_PROVIDER=lmstudio
LMSTUDIO_BASE_URL=http://localhost:1234/v1
LMSTUDIO_MODEL=your-local-model
```

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

### Troubleshooting Build Errors

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/START_HERE.md

This bash script provides a common solution for build errors in the ACE MCP Server project. It involves removing the existing node modules and build artifacts, then reinstalling dependencies and rebuilding the project.

```bash
# Delete and reinstall
rm -rf node_modules dist
npm install
npm run build
```

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

### Create Project Directory Structure (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/INITIALIZATION_REPORT.md

Creates the necessary directory structure for implementing LLM providers and their associated tests within the ace-mcp-server project.

```bash
mkdir -p src/llm src/llm/__tests__
```

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

### Troubleshooting API Key Format

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/LLM_PROVIDERS.md

Provides a bash command to check the format of an API key, specifically for DeepSeek. It echoes the API key and displays the first 10 characters to verify it starts with 'sk-', which is the expected format.

```bash
# Check your API key format
echo $DEEPSEEK_API_KEY | head -c 10
# Should start with 'sk-'
```

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

### Performance Tuning with DeepSeek Model and Timeout

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/LLM_PROVIDERS.md

Configures DeepSeek LLM provider for performance tuning. Examples show how to set the model to 'deepseek-chat' for faster responses or 'deepseek-reasoner' for better reasoning, along with adjusting the timeout values accordingly.

```env
# For faster responses
DEEPSEEK_MODEL=deepseek-chat
DEEPSEEK_TIMEOUT=30000

# For better reasoning (slower)
DEEPSEEK_MODEL=deepseek-reasoner
DEEPSEEK_TIMEOUT=120000
```

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

### Cursor AI Configuration for ACE MCP Server

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/START_HERE.md

This JSON configuration is used to set up the ACE MCP server within Cursor AI. It specifies the command to run the server, its arguments, and environment variables required for its operation. Remember to restart Cursor AI after updating this configuration.

```json
{
  "mcpServers": {
    "ace-context-engine": {
      "command": "node",
      "args": ["$HOME/code/perplexity/ace-mcp-server/dist/index.js"],
      "env": {
        "ACE_CONTEXT_DIR": "$HOME/code/perplexity/ace-mcp-server/contexts",
        "ACE_LOG_LEVEL": "info",
        "ACE_DEDUP_THRESHOLD": "0.85"
      }
    }
  }
}
```

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

### Verify Cursor Configuration Exists (Bash)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/ASSETS_CHECKLIST.md

Confirms that the necessary ACE context engine configuration is present in the Cursor AI user configuration file. This is required for integrating ACE with Cursor.

```bash
cat ~/.cursor/mcp.json | grep ace-context-engine
# Should show configuration
```

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

### Install Dependencies (npm)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/PROJECT_STATUS.md

Installs project dependencies for Ace MCP, including runtime libraries like 'openai', 'axios', 'zod', and development dependencies for testing with Jest and TypeScript.

```bash
npm install openai axios zod
npm install -D jest ts-jest @types/jest @types/axios
```

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

### Configure Production Environment Variables

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Sets up the `.env` file for production deployment. This includes critical settings such as the API bearer token, LLM provider and API key, Node environment, log level, and server IP/domain. Ensure sensitive information like API keys are kept secure.

```bash
# Copy example
cp .env.example .env

# Edit with your settings
nano .env
```

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

### Build and Run ACE MCP Server with Direct Docker Commands

Source: https://context7.com/angry-robot-deals/ace-mcp/llms.txt

This section details how to build and run the ACE MCP server using direct Docker commands, bypassing Docker Compose. It includes steps for building the Docker image, running the container with specific environment variables and volume mounts, creating necessary volumes, viewing logs, executing commands within the container, and stopping/removing the container and volumes.

```bash
# Build Docker image
docker build -t ace-mcp-server:latest .

# Run with environment variables
docker run -d \
  --name ace-mcp-server \
  -p 34301:34301 \
  -e LLM_PROVIDER=deepseek \
  -e DEEPSEEK_API_KEY=sk-your-key \
  -e DEEPSEEK_MODEL=deepseek-chat \
  -e API_BEARER_TOKEN=your-secure-token \
  -e ACE_SERVER_PORT=34301 \
  -e ACE_MAX_PLAYBOOK_SIZE=1000 \
  -v ace_contexts:/app/contexts \
  -v ace_logs:/app/logs \
  ace-mcp-server:latest

# Create volumes first
docker volume create ace_contexts
docker volume create ace_logs

# View logs
docker logs -f ace-mcp-server

# Execute test suite in container
docker exec ace-mcp-server npm test
docker exec ace-mcp-server node test-server.mjs

# Stop and remove
docker stop ace-mcp-server
docker rm ace-mcp-server

# Clean up volumes
docker volume rm ace_contexts ace_logs
```

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

### Example Usage: Smart Code Analysis (TypeScript)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_AUTO_SETUP.md

Illustrates the use of the `@ace_smart_reflect` command for analyzing code within Cursor AI. It provides a TypeScript function example and describes the types of feedback ACE offers.

```plaintext
```
@ace_smart_reflect 
```typescript
function quickSort(arr) {
  if (arr.length <= 1) return arr;
  const pivot = arr[0];
  const left = arr.slice(1).filter(x => x < pivot);
  const right = arr.slice(1).filter(x => x >= pivot);
  return [...quickSort(left), pivot, ...quickSort(right)];
}
```
```

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

### Enable Nginx Site and Reload Configuration

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DEPLOYMENT.md

Commands to enable the Nginx site configuration, disable the default site, test Nginx configuration syntax, and reload the Nginx service to apply changes.

```bash
ln -sf /etc/nginx/sites-available/ace-mcp /etc/nginx/sites-enabled/
rm -f /etc/nginx/sites-enabled/default
nginx -t
systemctl reload nginx
```

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

### Deploy ACE MCP Server with Docker Compose (Production)

Source: https://context7.com/angry-robot-deals/ace-mcp/llms.txt

This section provides instructions for deploying the ACE MCP Server and dashboard using Docker Compose in a production environment. It covers cloning the repository, configuring environment variables (including LLM provider and API keys), starting and stopping services, viewing logs, checking status, and rebuilding services.

```bash
# Clone repository
git clone https://github.com/Angry-Robot-Deals/ace-mcp.git
cd ace-mcp

# Configure environment
cp .env.example .env
nano .env  # Set LLM_PROVIDER, API keys, and tokens

# Start services
docker-compose up -d

# View logs
docker-compose logs -f ace-server
docker-compose logs -f ace-dashboard

# Check status
curl http://localhost:34301/health
curl http://localhost:34300  # Dashboard

# Stop services
docker-compose down

# Rebuild services
docker-compose build --no-cache
docker-compose up -d
```

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

### TypeScript: ACE Structured Bullet Example

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/intro/DESCRIPTION.md

An example of a structured bullet object used within the ACE framework. This format is more compact and efficient than plain text prose, leading to reduced storage requirements and faster processing.

```typescript
{
  id: "b001",
  content: "Use Zod for API validation",
  helpful_count: 12
} // ~100 bytes vs 1000+ byte prose

```

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

### Run ACE MCP Development Environment with Docker Compose

Source: https://context7.com/angry-robot-deals/ace-mcp/llms.txt

Instructions for running the ACE MCP development environment using Docker Compose, which includes live reloading. This covers starting the environment, viewing live logs, executing commands within the container (e.g., running tests), and stopping the development environment.

```bash
# Start development environment
docker-compose -f docker-compose.dev.yml up -d

# View live logs
docker-compose -f docker-compose.dev.yml logs -f

# Execute commands in container
docker-compose -f docker-compose.dev.yml exec ace-server npm test

# Stop development environment
docker-compose -f docker-compose.dev.yml down
```

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

### Example Usage: Smart Code Generation (Command Prompt)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/setup/CURSOR_AI_AUTO_SETUP.md

Demonstrates how to use the `@ace_smart_generate` command for creating code in Cursor AI. It shows the difference between a regular request and an enhanced request, highlighting the automatic improvements ACE provides.

```plaintext
Instead of a regular request:
```
Create a function to sort an array
```

Use:
```
@ace_smart_generate create a function to sort an array
```
```

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

### Read Main Rule and Tasks Files

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/custom_modes/plan_instructions.md

This code snippet demonstrates how to read the main rule file and the tasks file. It utilizes a 'read_file' function, specifying the target file paths and indicating that the entire file content should be read. This is a foundational step in the planning mode.

```javascript
read_file({
  target_file: ".cursor/rules/isolation_rules/main.mdc",
  should_read_entire_file: true
})

read_file({
  target_file: "tasks.md",
  should_read_entire_file: true
})
```

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

### Describe Test Cases (TypeScript)

Source: https://github.com/angry-robot-deals/ace-mcp/blob/main/docs/CONTRIBUTING.md

Example of how to structure test cases in TypeScript using `describe` and `it` blocks. Demonstrates best practices for naming test suites and individual tests, including handling specific scenarios like adding bullets to sections and managing empty playbooks.

```typescript
describe('Playbook', () => {
  it('should add bullet to correct section', () => {
    // test implementation
  });
  
  it('should handle empty playbook gracefully', () => {
    // test implementation
  });
});
```