### Local Development Setup

Source: https://github.com/agent0ai/agent-zero/blob/main/knowledge/main/about/setup-and-deployment.md

Sets up a Python virtual environment, installs dependencies, and runs the Agent Zero UI for local development. Ensure you are in the project root directory.

```bash
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install -r requirements2.txt
python run_ui.py
```

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

### Example First Chat Prompt

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/quickstart.md

Use this prompt to get Agent Zero to create a short plan for organizing your project notes.

```text
Create a short plan for organizing my project notes.
```

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

### Set Up Virtual Environment and Install Dependencies

Source: https://github.com/agent0ai/agent-zero/blob/main/AGENTS.md

Commands to create a virtual environment, activate it, and install project dependencies from requirement files. Use this when encountering pip installation issues.

```bash
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install -r requirements2.txt
```

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

### Install Qwen Code

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs the Qwen Code CLI globally using npm. This command ensures you get the latest version.

```bash
npm install -g @qwen-code/qwen-code@latest
qwen
```

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

### A0 CLI Manual Installation with uv

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_a0_connector/skills/setup-a0-cli/SKILL.md

If the direct installer URL is blocked, use `uv tool install` for a manual installation. This method allows for specifying package sources like local checkouts or internal Git hosts.

```bash
uv tool install --upgrade <package-spec>
```

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

### Install Docker on Debian/Ubuntu

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/vps-deployment.md

Installs Docker Engine, CLI, and plugins on Debian-based systems. Ensures Docker is enabled and started.

```bash
# Update package index
apt-get update

# Install prerequisites
apt-get install -y ca-certificates curl gnupg

# Add Docker's official GPG key
install -m 0755 -d /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | gpg --dearmor -o /etc/apt/keyrings/docker.gpg
chmod a+r /etc/apt/keyrings/docker.gpg

# Set up repository
echo   "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu   $(. /etc/os-release && echo "$VERSION_CODENAME") stable" |   tee /etc/apt/sources.list.d/docker.list > /dev/null

# Install Docker
apt-get update
apt-get install -y docker-ce docker-ce-cli containerd.io docker-compose-plugin

# Start and enable Docker
systemctl enable docker
systemctl start docker
```

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

### Install Docker on AlmaLinux/Rocky/CentOS/RHEL

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/vps-deployment.md

Installs Docker Engine, CLI, and plugins on RHEL-based systems. Ensures Docker is enabled and started.

```bash
# Install required packages
dnf -y install dnf-plugins-core

# Add Docker repository (use CentOS repo for AlmaLinux/Rocky)
dnf config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo

# Install Docker
dnf -y install docker-ce docker-ce-cli containerd.io docker-compose-plugin

# Start and enable Docker
systemctl enable docker
systemctl start docker
```

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

### Install OpenCode (NPM Alternative)

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs the OpenCode CLI globally using npm. This is an alternative installation method.

```bash
npm i -g opencode
opencode
```

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

### Fallback A0 CLI Install via uv tool install (Public Git)

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_a0_connector/skills/setup-a0-cli/SKILL.md

If the direct installer URL is blocked or unavailable, use this command to install the A0 CLI using `uv tool install` with a public Git repository.

```bash
uv tool install --upgrade git+https://github.com/agent0ai/a0-connector
```

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

### Fallback A0 CLI Install via uv tool install (Local Checkout)

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_a0_connector/skills/setup-a0-cli/SKILL.md

Use this command to install the A0 CLI from a local checkout of the repository using `uv tool install`.

```bash
uv tool install --upgrade /path/to/a0-connector
```

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

### Install Docker using Convenience Script

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/vps-deployment.md

Installs Docker using a convenience script. Note: May not work on all distributions.

```bash
curl -fsSL https://get.docker.com -o get-docker.sh
sh get-docker.sh
systemctl enable docker
systemctl start docker
```

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

### Get Files Example

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/external/api-examples.html

Retrieves content of specified files from the server. Supports text files and indicates binary files.

```javascript
// Basic file retrieval
async function getFiles(filePaths) {
  try {
    const response = await fetch('${url}/api/api_files_get', {
      method: 'POST',
      headers: {
        'Content-Type': 'application/json',
        'X-API-KEY': '${token}'
      },
      body: JSON.stringify({ paths: filePaths })
    });
    const data = await response.json();
    if (response.ok) {
      console.log('✅ Files retrieved successfully!');
      console.log('Retrieved files:', Object.keys(data));
      // Convert base64 back to text for display
      for (const [filename, base64Content] of Object.entries(data)) {
        try {
          const textContent = atob(base64Content);
          console.log(`
${filename}: ${textContent.substring(0, 100)}...`);
        } catch (e) {
          console.log(`
${filename}: Binary file (${base64Content.length} chars)`);
        }
      }
      return data;
    } else {
      console.error('❌ Error:', data.error);
      return null;
    }
  } catch (error) {
    console.error('❌ Request failed:', error);
    return null;
  }
}

// Example 1: Get specific files
const filePaths = [
  "/a0/usr/uploads/document.txt",
  "/a0/usr/uploads/data.json"
];
getFiles(filePaths);

// Example 2: Complete attachment workflow
async function attachmentWorkflow() {
  // Step 1: Send message with attachments
  const messageResponse = await fetch('${url}/api/api_message', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'X-API-KEY': '${token}'
    },
    body: JSON.stringify({
      message: "Please analyze this file",
      attachments: [
        {
          filename: "test.txt",
          base64: btoa("Hello, this is tes
```

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

### Fallback A0 CLI Install via uv tool install (Internal Mirror)

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_a0_connector/skills/setup-a0-cli/SKILL.md

Install the A0 CLI from an internal Git mirror using `uv tool install`. Ensure the Git URL is accessible.

```bash
uv tool install --upgrade git+ssh://git.example.com/team/a0-connector.git
```

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

### Install OpenAI Codex

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs the OpenAI Codex CLI using npm and then launches it. Ensure Node.js and npm are installed.

```bash
npm i -g @openai/codex
codex
```

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

### Install OpenClaw

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs OpenClaw globally using npm and then initiates its onboarding process, including installing a daemon. Requires Node.js 24 or 22.14+.

```bash
npm install -g openclaw@latest
openclaw onboard --install-daemon
```

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

### Load Memories Example

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_memory/prompts/agent.system.tool.memory.md

Example of how to structure a tool call to load memories based on a query, similarity threshold, and limit.

```json
{
  "thoughts": ["I should search memory for relevant prior guidance."],
  "headline": "Loading related memories",
  "tool_name": "memory_load",
  "tool_args": {
    "query": "tool argument format",
    "threshold": 0.7,
    "limit": 3
  }
}
```

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

### API Files Get Example

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/external/api-examples.html

Retrieves a list of files from the API based on provided paths. The response is a dictionary mapping filenames to their content.

```javascript
// API files get example
async function getFiles(filePaths) {
  const response = await fetch(`${url}/api/api_files`, {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'X-API-KEY': '${token}'
    },
    body: JSON.stringify({ paths: filePaths })
  });
  const data = await response.json();
  if (response.ok) {
    console.log('Files retrieved successfully:');
    return data.files;
  } else {
    console.error('Error retrieving files:', data.error);
    return null;
  }
}

// Example usage: retrieve a specific file
// const retrievedFiles = await getFiles(["/a0/usr/uploads/test.txt"]);
// if (retrievedFiles && retrievedFiles["test.txt"]) {
//   const originalContent = atob(retrievedFiles["test.txt"]);
//   console.log('Retrieved content:', originalContent);
// }

```

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

### Install OpenCode

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs OpenCode using a curl script and then sources your bashrc. This is a convenient way to install the CLI.

```bash
curl -sL https://opencode.ai/install | bash
opencode
```

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

### Setup A0 CLI Connector

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/a0-cli-connector.md

Provide this single line to another agent if they are assisting with setup, instructing them to use the setup-a0-cli Skill.

```text
Set up the A0 CLI connector for Agent Zero on this machine using the setup-a0-cli Skill.
```

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

### Install Plugin from Git URL via HTTP API

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-manage-plugin/SKILL.md

Installs a plugin by cloning it from a Git repository URL. This method handles the full installation pipeline, including validation, placement, hook execution, and cache clearing.

```python
# (after authentication setup above)
resp = s.post(
    f"{BASE}/api/plugins/_plugin_installer/plugin_install",
    json={
        "action": "install_git",
        "git_url": "https://github.com/<user>/<plugin-repo>",
        # "git_token": "<token>",    # optional, for private repos
        # "plugin_name": "override"  # optional, override directory name
    },
    headers={"X-CSRF-Token": token, "Origin": ORIGIN},
    timeout=120,
)
print(resp.json())
```

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

### Install Plugin from Git URL

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-manage-plugin/SKILL.md

Installs a plugin from a Git repository URL. This is the preferred method for programmatic use. It handles cloning, validation, installation, and cache clearing.

```APIDOC
## POST /api/plugins/_plugin_installer/plugin_install

### Description
Installs a plugin from a Git repository URL. This method triggers a full installation pipeline including cloning, validation, placing the plugin in `usr/plugins/`, running the `install` hook, and clearing the plugin cache.

### Method
POST

### Endpoint
/api/plugins/_plugin_installer/plugin_install

### Parameters
#### Request Body
- **action** (string) - Required - Must be `"install_git"`.
- **git_url** (string) - Required - The URL of the Git repository containing the plugin.
- **git_token** (string) - Optional - A token for accessing private Git repositories.
- **plugin_name** (string) - Optional - Overrides the directory name for the plugin.
```

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

### Install WhatsApp Bridge Dependencies

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_whatsapp_integration/README.md

Commands to navigate to the bridge directory and install required production dependencies.

```bash
cd plugins/_whatsapp_integration/whatsapp-bridge
npm install --production
```

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

### Install Ollama on Linux

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/installation.md

Run this script to install Ollama on Linux systems.

```bash
curl -fsSL https://ollama.com/install.sh | sh
```

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

### Install Plugin from ZIP File

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-manage-plugin/SKILL.md

Installs a plugin by uploading a ZIP file. This method is also available via the HTTP API.

```APIDOC
## POST /api/plugins/_plugin_installer/plugin_install

### Description
Installs a plugin by uploading a ZIP file. The file is processed by the framework to install the plugin.

### Method
POST

### Endpoint
/api/plugins/_plugin_installer/plugin_install

### Parameters
#### Request Body
- **action** (string) - Required - Must be `"install_zip"`.
- **plugin_file** (file) - Required - The ZIP file containing the plugin.
```

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

### Start Computer Use Session

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_a0_connector/skills/host-computer-use/SKILL.md

Initiates a session with the computer use tool. This is the first action to call when starting a sequence of computer interactions.

```json
{
  "tool_name": "computer_use_remote",
  "tool_args": {
    "action": "start_session"
  }
}
```

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

### Install A0 CLI on Windows (PowerShell)

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/a0-cli-connector.md

Installs the A0 CLI connector using a PowerShell script. Run this on your host machine.

```powershell
irm https://cli.agent-zero.ai/install.ps1 | iex
```

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

### Basic SKILL.md Content Example

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/developer/contributing-skills.md

An example of the essential fields for a SKILL.md file, including name, description, version, author, and tags.

```markdown
---
name: "my-awesome-skill"
description: "Helps with [specific task] when [specific situation]"
version: "1.0.0"
author: "Your Name"
tags: ["category"]
---

# My Awesome Skill

## When to Use

Use this skill when you need to [specific task].

## Instructions

1. First, do this...
2. Then, do that...
3. Finally, verify by...

## Examples

### Example: Basic Case

[Show a complete example]

```

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

### Install Gemini CLI

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs the Gemini CLI using npm. This command assumes npm is available in your environment.

```bash
npm install -g @google/gemini-cli
gemini
```

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

### Install A0 CLI on macOS/Linux

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/a0-cli-connector.md

Installs the A0 CLI connector using a curl script. Run this on your host machine.

```bash
curl -LsSf https://cli.agent-zero.ai/install.sh | sh
```

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

### Example API Endpoint Implementation

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-development/SKILL.md

An example demonstrating how to implement a custom API endpoint by extending the ApiHandler class and defining its behavior.

```APIDOC
## Example: API Endpoint

### Description
This example shows a basic API endpoint that accepts GET and POST requests. It processes an input parameter and returns a result along with the context ID.

### Method
- `GET`, `POST`

### Endpoint
Auto-discovered based on filename (e.g., `my_endpoint.py` maps to `/api/my_endpoint`). The base path is typically `/a0/api/`.

### Parameters
#### Request Body (for POST)
- **param** (string) - Optional - The input parameter to process. Defaults to "default" if not provided.
- **context** (string) - Required - The ID of the agent context to use.

### Request Example
```json
{
  "param": "some_value",
  "context": "agent_context_id"
}
```

### Response
#### Success Response (200)
- **result** (string) - The processed result string.
- **context** (string) - The ID of the agent context used.

#### Response Example
```json
{
  "result": "processed some_value",
  "context": "agent_context_id"
}
```
```

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

### Install Plugin from ZIP File via HTTP API

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-manage-plugin/SKILL.md

Installs a plugin by uploading a ZIP archive. The plugin file is sent as multipart/form-data.

```python
# (after authentication setup above)
with open("plugin.zip", "rb") as f:
    resp = s.post(
        f"{BASE}/api/plugins/_plugin_installer/plugin_install",
        data={"action": "install_zip"},
        files={"plugin_file": f},
        headers={"X-CSRF-Token": token, "Origin": ORIGIN},
    )
print(resp.json())
```

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

### List Installed Plugins Locally

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-manage-plugin/SKILL.md

Lists the names of plugins currently installed in the Agent Zero environment. This can be done via the command line.

```bash
ls /a0/usr/plugins/

```

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

### Example Project Creation Prompt

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/quickstart.md

Use this prompt to ask Agent Zero to help create a project for a repository and write instructions for it.

```text
Help me create a project for this repository and write good instructions for it.
```

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

### Example Project Instructions

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/projects.md

Define project-specific behavior and guidelines for Agent Zero. This example focuses on documentation tasks, emphasizing clear explanations, user-facing content, and specific file handling rules.

```markdown
You are working inside the Docs Example Workspace.

Use this project for small documentation examples and user-facing guidance.

When this project is active:
- Explain steps in plain language before technical detail.
- Prefer screenshots, checklists, and concrete examples.
- Keep generated files inside this project unless I ask otherwise.
- Ask before using credentials, private data, or external accounts.
- When editing docs, focus on what the user sees and what they should do next.
```

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

### Agent Profile agent.yaml Example

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-development/SKILL.md

Example of the agent.yaml file format for defining agent profiles. It includes fields for title, description, and context, which guide the agent's specialization and usage.

```yaml
title: Developer
description: Agent specialized in complex software development.
context: Use this agent for software development tasks, including writing code,
debugging, refactoring, and architectural design.
```

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

### Example Tool Implementation

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-development/SKILL.md

Demonstrates how to create a custom tool by inheriting from `Tool` and implementing the `execute` method to process input and return a response.

```python
# my_tool.py
from helpers.tool import Tool, Response

class MyTool(Tool):
    async def execute(self, **kwargs):
        # Get arguments — kwargs contains the tool_args from the agent's JSON
        input_data = kwargs.get("input", "")

        # Do something
        result = f"Processed: {input_data}"

        # Return response
        return Response(
            message=result,        # Shown to the agent
            break_loop=False,      # Don't stop the agent loop
        )
```

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

### POST /agent/tools/example_tool

Source: https://github.com/agent0ai/agent-zero/blob/main/agents/_example/prompts/agent.system.tool.example_tool.md

Executes the example tool to verify system functionality and tool integration.

```APIDOC
## POST /agent/tools/example_tool

### Description
Executes the example_tool to test system functionality. This tool is automatically included in the system prompt.

### Method
POST

### Endpoint
/agent/tools/example_tool

### Parameters
#### Request Body
- **thoughts** (array) - Required - A list of strings representing the agent's reasoning process.
- **headline** (string) - Required - A brief summary of the action.
- **tool_name** (string) - Required - The name of the tool to execute (must be "example_tool").
- **tool_args** (object) - Required - Arguments for the tool.
  - **test_input** (string) - Required - The input string to be processed by the tool.

### Request Example
{
    "thoughts": ["Let's test the example tool..."],
    "headline": "Testing example tool",
    "tool_name": "example_tool",
    "tool_args": {
        "test_input": "XYZ"
    }
}

### Response
#### Success Response (200)
- **status** (string) - Indicates the execution result.

#### Response Example
{
    "status": "success",
    "output": "Processed XYZ"
}
```

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

### Retrieve Logs using GET Request

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/external/api-examples.html

This example demonstrates how to retrieve logs using a GET request. It requires a `context_id` and optionally accepts a `length` parameter to specify the number of log items to retrieve.

```javascript
// Get logs using GET request
async function getLogsGET(contextId, length = 50) {
  try {
    const params = new URLSearchParams({ context_id: contextId, length: length.toString() });
    const response = await fetch('${url}/api/api_log_get?' + params, {
      method: 'GET',
      headers: {
        'X-API-KEY': '${token}'
      }
    });
    const data = await response.json();
    if (response.ok) {
      console.log('✅ Logs retrieved successfully!');
      console.log('Total items:', data.log.total_items);
      console.log('Returned items:', data.log.returned_items);
      console.log('Log items:', data.log.items);
      return data;
    } else {
      console.error('❌ Error:', data.error);
      return null;
    }
  } catch (error) {
    console.error('❌ Request failed:', error);
    return null;
  }
}

// Example usage
getLogsGET('ctx_abc123', 20);
```

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

### Get Logs Example

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/external/api-examples.html

Retrieves log entries for a given context ID. Specify the desired number of log entries to fetch.

```javascript
async function getLogsPOST(contextId, length) {
  try {
    const response = await fetch(`${url}/api/api_logs`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json', 'X-API-KEY': '${token}' },
      body: JSON.stringify({ context_id: contextId, length: length })
    });
    const data = await response.json();
    if (response.ok) {
      console.log('✅ Logs retrieved successfully!');
      console.log('Context ID:', data.context_id);
      console.log('Log GUID:', data.log.guid);
      console.log('Total items:', data.log.total_items);
      console.log('Returned items:', data.log.returned_items);
      console.log('Start position:', data.log.start_position);
      console.log('Progress:', data.log.progress);
      console.log('Log items:', data.log.items);
      return data;
    } else {
      console.error('❌ Error:', data.error);
      return null;
    }
  } catch (error) {
    console.error('❌ Request failed:', error);
    return null;
  }
}
// Example usage - get latest 10 log entries
getLogsPOST('ctx_abc123', 10);
```

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

### Example @extensible Path Mapping

Source: https://github.com/agent0ai/agent-zero/blob/main/skills/a0-development/SKILL.md

Illustrates how a specific function maps to the generated extension point directories.

```text
_functions/agent/Agent/handle_exception/start
```

```text
_functions/agent/Agent/handle_exception/end
```

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

### API Reset Example

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/external/api-examples.html

Sends a request to reset the agent's state. This is useful for starting a fresh conversation or clearing previous context.

```javascript
// API reset example
async function resetAgent() {
  const response = await fetch(`${url}/api/api_reset`, {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'X-API-KEY': '${token}'
    },
    body: JSON.stringify({})
  });
  const data = await response.json();
  if (response.ok) {
    console.log('Agent reset successfully!');
  } else {
    console.error('Error resetting agent:', data.error);
  }
}

// Call the function
resetAgent();

```

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

### Setup Multiple Agent Zero Instances

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/vps-deployment.md

Demonstrates the directory structure and basic commands for deploying multiple Agent Zero instances on a single server, each with unique ports and data directories.

```bash
# Instance 1: a0-primary on port 50080
mkdir -p /opt/a0-primary
# ... create .env, run container on port 50080

# Instance 2: a0-dev on port 50081  
mkdir -p /opt/a0-dev
# ... create .env, run container on port 50081

# Instance 3: a0-backup on port 50082
mkdir -p /opt/a0-backup
# ... create .env, run container on port 50082
```

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

### Direct Docker Run for Agent Zero

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/installation.md

Run the Agent Zero Docker image directly to start the service. This is an alternative to the install scripts.

```bash
docker run -p 80:80 agent0ai/agent-zero
```

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

### Skill Directory Structure with Supporting Files

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/developer/contributing-skills.md

Example of a skill's directory structure, including the required SKILL.md and optional supporting scripts or template files.

```bash
my-awesome-skill/
├── SKILL.md           # Required
├── helper.py          # Optional Python script
├── setup.sh           # Optional shell script
└── templates/         # Optional templates folder
    └── config.json

```

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

### Project Instructions Example

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/usage.md

Define specific instructions for Agent Zero within a project context, focusing on communication style and file handling.

```markdown
When this project is active:
- Explain steps in plain language before technical detail.
- Prefer screenshots, checklists, and concrete examples.
- Keep generated files inside this project unless I ask otherwise.
- Ask before using credentials, private data, or external accounts.
```

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

### Invoke example_tool via JSON

Source: https://github.com/agent0ai/agent-zero/blob/main/agents/_example/prompts/agent.system.tool.example_tool.md

Demonstrates the JSON structure required to invoke the example_tool. It includes a thoughts array, a headline, the tool name, and the specific arguments required by the tool.

```json
{
    "thoughts": [
        "Let's test the example tool..."
    ],
    "headline": "Testing example tool",
    "tool_name": "example_tool",
    "tool_args": {
        "test_input": "XYZ"
    }
}
```

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

### Run Agent Zero with Docker

Source: https://github.com/agent0ai/agent-zero/blob/main/README.md

If Docker is already installed, run this command to start Agent Zero. It maps port 80 and mounts a volume for user data.

```bash
docker run -p 80:80 -v a0_usr:/a0/usr agent0ai/agent-zero
```

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

### Importing Example Store Component (JavaScript)

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/_examples/_example-component.html

Demonstrates how to import a store component from a specified path within the project. This is a common pattern for managing application state or shared logic across components.

```javascript
import { store } from "/components/_examples/_example-store.js";
```

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

### Execute Node.js Code to Get Current Directory

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_code_execution/prompts/agent.system.tool.code_exe.md

Execute Javascript code using Node.js to perform operations like retrieving the current working directory. Ensure Node.js is installed in the execution environment.

```json
{
    "thoughts": [
        "Need to do...",
        "I can use...",
        "Then I can...",
    ],
    "headline": "Executing Javascript code to check current directory",
    "tool_name": "code_execution_tool",
    "tool_args": {
        "runtime": "nodejs",
        "session": 0,
        "reset": false,
        "code": "console.log(process.cwd());",
    }
}
```

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

### Example Profile Creation Prompt

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/agent-profiles.md

Provide a practical description of the desired profile's capabilities, target audience, tone, and output format.

```text
This profile should help me plan YouTube scripts for technical demos. It should
ask for the target audience, keep the tone simple, and suggest a visual outline.

```

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

### Install Claude Code (Native Installer)

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs Claude Code using a recommended native installer script. After installation, source your bashrc to make the command available.

```bash
curl -fsSL https://claude.ai/install.sh | bash
source ~/.bashrc
claude
```

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

### Install Ollama on macOS using Homebrew

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/installation.md

Use this command to install Ollama on macOS if you have Homebrew installed.

```bash
brew install ollama
```

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

### Styling Example Component (CSS)

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/_examples/_example-component.html

Provides CSS rules for styling an element with the ID 'example-component'. This snippet shows how to apply basic width styling to ensure the component occupies full available width.

```css
#example-component { width: 100%; }
```

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

### Create Agent Zero Directory Structure

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/setup/vps-deployment.md

Sets up the necessary directories for Agent Zero instance data and logs.

```bash
# Choose your installation path
A0_NAME="a0-instance"  # Change this to your instance name
A0_PATH="/opt/${A0_NAME}"

# Create directories
mkdir -p ${A0_PATH}
mkdir -p ${A0_PATH}/work_dir
mkdir -p ${A0_PATH}/memory
mkdir -p ${A0_PATH}/logs
```

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

### Install Claude Code (NPM Alternative)

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs the Claude Code CLI as an npm package. This is an alternative to the native installer.

```bash
npm install -g @anthropic-ai/claude-code
claude
```

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

### API Project Management Example

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/external/api-examples.html

Demonstrates activating a project by name in the first message and then continuing the conversation within that project's context. Subsequent messages do not need the project name.

```javascript
// Project usage example
async function sendMessageWithProject() {
  try {
    // First message - activate project
    const response = await fetch('${url}/api/api_message', {
      method: 'POST',
      headers: {
        'Content-Type': 'application/json',
        'X-API-KEY': '${token}'
      },
      body: JSON.stringify({
        message: "Analyze the project structure",
        project_name: "my-web-app" // Activates this project
      })
    });
    const data = await response.json();
    if (response.ok) {
      console.log('✅ Project activated!');
      console.log('Context ID:', data.context_id);
      console.log('Response:', data.response);
      // Continue conversation - project already set
      const followUp = await fetch('${url}/api/api_message', {
        method: 'POST',
        headers: {
          'Content-Type': 'application/json',
          'X-API-KEY': '${token}'
        },
        body: JSON.stringify({
          context_id: data.context_id,
          message: "What files are in the project?" // Do NOT include project_name here - already set on first message
        })
      });
      const followUpData = await followUp.json();
      console.log('Follow-up response:', followUpData.response);
      return followUpData;
    } else {
      console.error('❌ Error:', data.error);
      return null;
    }
  } catch (error) {
    console.error('❌ Request failed:', error);
    return null;
  }
}

// Call the function
sendMessageWithProject();

```

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

### Example MCP Servers Configuration JSON

Source: https://github.com/agent0ai/agent-zero/blob/main/webui/components/settings/mcp/client/example.html

This JSON object demonstrates the structure for configuring MCP servers. It includes examples of local servers defined by 'command' and 'args', and remote servers defined by 'url' and 'headers'. Options like 'disabled', 'description', 'init_timeout', and 'tool_timeout' are also illustrated.

```json
{
  "mcpServers": {
    "sqlite": {
      "command": "uvx",
      "args": [
        "mcp-server-sqlite",
        "--db-path",
        "/root/db.sqlite"
      ],
      "init_timeout": 10,
      "tool_timeout": 200
    },
    "sequential-thinking": {
      "disabled": true,
      "command": "npx",
      "args": [
        "--yes",
        "--package",
        "@modelcontextprotocol/server-sequential-thinking",
        "mcp-server-sequential-thinking"
      ]
    },
    "deep-wiki": {
      "description": "Use this MCP to analyze github repositories",
      "url": "https://mcp.deepwiki.com/mcp",
      "type": "streamable-http"
    }
  }
}
```

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

### Install OpenHands CLI

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_desktop/assets/desktop/README.md

Installs OpenHands CLI using pip with uv as the package manager. It specifies Python 3.12 for the installation.

```bash
python3 -m pip install uv
uv tool install openhands --python 3.12
openhands
```

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

### Install Playwright Browsers for Local Development

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/troubleshooting.md

Run this command from the project root after installing Python requirements to install the Chromium browser for local development checkouts.

```bash
PLAYWRIGHT_BROWSERS_PATH=tmp/playwright playwright install chromium
```

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

### Create Presentation and Check Slides in Desktop

Source: https://github.com/agent0ai/agent-zero/blob/main/docs/guides/desktop.md

Use this prompt to create a short Impress presentation and then use the Desktop to verify the slide appearance.

```text
Create a short Impress deck, then use Desktop to check that the slides look clean.
```

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

### A0 CLI Installation within Container

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_a0_connector/skills/setup-a0-cli/SKILL.md

The A0 CLI must be installed on the host machine, not within a Docker container. If you are inside a container, exit and run the host installer.

```bash
exit
```

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

### Example Output Format

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_memory/prompts/memory.memories_filter.sys.md

The AI should respond with a JSON array of integers representing the indices of relevant memories. No other text or formatting should be included.

```json
[0, 2]
```

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

### Import Plugin Installer Store JavaScript

Source: https://github.com/agent0ai/agent-zero/blob/main/plugins/_plugin_installer/webui/install-git.html

Imports the store object from the plugin installer's web UI JavaScript module. This is essential for managing the state of plugin installations within the application.

```javascript
import { store } from "/plugins/_plugin_installer/webui/pluginInstallStore.js";
```