`Agent Zero`

Documentation:

Introduction • Installation • How to update
Development Setup • Usage

Or see DeepWiki generated documentation:

🚨 AGENT ZERO SKILLS 🚨

Skills System - portable, structured agent capabilities using the open SKILL.md standard (compatible with Claude Code, Codex and more).

Plus: Git-based Projects with authentication for public/private repositories - clone codebases directly into isolated workspaces.

See Usage Guide and Projects Tutorial to get started.

A personal, organic agentic framework that grows and learns with you

Agent Zero is not a predefined agentic framework. It is designed to be dynamic, organically growing, and learning as you use it.
Agent Zero is fully transparent, readable, comprehensible, customizable, and interactive.
Agent Zero uses the computer as a tool to accomplish its (your) tasks.

⚙️ Installation

Click to open a video to learn how to install Agent Zero:

A detailed setup guide for Windows, macOS, and Linux with a video can be found in the Agent Zero Documentation at this page.

⚡ Quick Start

# Pull and run with Docker

docker pull agent0ai/agent-zero
docker run -p 50001:80 agent0ai/agent-zero

# Visit http://localhost:50001 to start

💡 Key Features

General-purpose Assistant

Agent Zero is not pre-programmed for specific tasks (but can be). It is meant to be a general-purpose personal assistant. Give it a task, and it will gather information, execute commands and code, cooperate with other agent instances, and do its best to accomplish it.
It has a persistent memory, allowing it to memorize previous solutions, code, facts, instructions, etc., to solve tasks faster and more reliably in the future.

Agent 0 Working

Computer as a Tool

Agent Zero uses the operating system as a tool to accomplish its tasks. It has no single-purpose tools pre-programmed. Instead, it can write its own code and use the terminal to create and use its own tools as needed.
The only default tools in its arsenal are online search, memory features, communication (with the user and other agents), and code/terminal execution. Everything else is created by the agent itself or can be extended by the user.
Tool usage functionality has been developed from scratch to be the most compatible and reliable, even with very small models.
Default Tools: Agent Zero includes tools like knowledge, code execution, and communication.
Creating Custom Tools: Extend Agent Zero's functionality by creating your own custom tools.
Skills (SKILL.md Standard): Skills are contextual expertise loaded dynamically when relevant. They use the open SKILL.md standard (developed by Anthropic), making them compatible with Claude Code, Cursor, Goose, OpenAI Codex CLI, and GitHub Copilot.

Multi-agent Cooperation

Every agent has a superior agent giving it tasks and instructions. Every agent then reports back to its superior.
In the case of the first agent in the chain (Agent 0), the superior is the human user; the agent sees no difference.
Every agent can create its subordinate agent to help break down and solve subtasks. This helps all agents keep their context clean and focused.

Multi-agent

Completely Customizable and Extensible

Almost nothing in this framework is hard-coded. Nothing is hidden. Everything can be extended or changed by the user.
The whole behavior is defined by a system prompt in the prompts/default/agent.system.md file. Change this prompt and change the framework dramatically.
The framework does not guide or limit the agent in any way. There are no hard-coded rails that agents have to follow.
Every prompt, every small message template sent to the agent in its communication loop can be found in the prompts/ folder and changed.
Every default tool can be found in the python/tools/ folder and changed or copied to create new predefined tools.
Automated configuration via A0_SET_ environment variables for deployment automation and easy setup.

Prompts

Communication is Key

Give your agent a proper system prompt and instructions, and it can do miracles.
Agents can communicate with their superiors and subordinates, asking questions, giving instructions, and providing guidance. Instruct your agents in the system prompt on how to communicate effectively.
The terminal interface is real-time streamed and interactive. You can stop and intervene at any point. If you see your agent heading in the wrong direction, just stop and tell it right away.
There is a lot of freedom in this framework. You can instruct your agents to regularly report back to superiors asking for permission to continue. You can instruct them to use point-scoring systems when deciding when to delegate subtasks. Superiors can double-check subordinates' results and dispute. The possibilities are endless.

🚀 Real-world use cases

Financial Analysis & Charting - "Find last month's Bitcoin/USD price trend, correlate with major cryptocurrency news events, generate annotated chart with highlighted key dates"
Excel Automation Pipeline - "Scan incoming directory for financial spreadsheets, validate and clean data, consolidate from multiple sources, generate executive reports with flagged anomalies"
API Integration Without Code - "Use this Google Gemini API snippet to generate product images, remember the integration for future use" - agent learns and stores the solution in memory
Automated Server Monitoring - "Check server status every 30 minutes: CPU usage, disk space, memory. Alert if metrics exceed thresholds" (scheduled task with project-scoped credentials)
Multi-Client Project Isolation - Separate projects for each client with isolated memory, custom instructions, and dedicated secrets - prevents context bleed across sensitive work

🐳 Fully Dockerized, with Speech-to-Text and TTS

Settings

Customizable settings allow users to tailor the agent's behavior and responses to their needs.
The Web UI output is very clean, fluid, colorful, readable, and interactive; nothing is hidden.
You can load or save chats directly within the Web UI.
The same output you see in the terminal is automatically saved to an HTML file in logs/ folder for every session.

Time example

Agent output is streamed in real-time, allowing users to read along and intervene at any time.
No coding is required; only prompting and communication skills are necessary.
With a solid system prompt, the framework is reliable even with small models, including precise tool usage.

👀 Keep in Mind

Agent Zero Can Be Dangerous!

With proper instruction, Agent Zero is capable of many things, even potentially dangerous actions concerning your computer, data, or accounts. Always run Agent Zero in an isolated environment (like Docker) and be careful what you wish for.

Agent Zero Is Prompt-based.

The whole framework is guided by the prompts/ folder. Agent guidelines, tool instructions, messages, utility AI functions, it's all there.

📚 Read the Documentation

Page	Description
Installation	Installation, setup and configuration
Usage	Basic and advanced usage
Guides	Step-by-step guides: Usage, Projects, API Integration, MCP Setup, A2A Setup
Development Setup	Development and customization
WebSocket Infrastructure	Real-time WebSocket handlers, client APIs, filtering semantics, envelopes
Extensions	Extending Agent Zero
Connectivity	External API endpoints, MCP server connections, A2A protocol
Architecture	System design and components
Contributing	How to contribute
Troubleshooting	Common issues and their solutions

🎯 Changelog

v0.9.8 - Skills, UI Redesign & Git projects

Release video

Skills
- Skills System replacing the legacy Instruments with a new SKILL.md standard for structured, portable agent capabilities.
- Built-in skills, and UI support for importing and listing skills
Real-time WebSocket infrastructure replacing the polling-based approach for UI state synchronization
UI Redesign
- Process groups to visually group agent actions with expand/collapse support
- Timestamps, steps count and execution time with tool-specific badges
- Step detail modals with key-value and raw JSON display
- Collapsible responses with show more/less and copy buttons on code blocks and tables
- Message queue system allowing users to queue messages while the agent is still processing
- In-browser file editor for viewing and editing files without leaving the UI
- Welcome screen redesign with info and warning banners for connection security, missing API keys, and system resources
- Scheduler redesign with standalone modal, separate task list, detail and editor components, and project support
- Smooth response rendering and scroll stabilization across chat, terminals, and image viewer
- Chat width setting and reworked preferences panel
- Image viewer improvements with scroll support and expanded viewer
- Redesigned sidebar with reusable dropdown component and streamlined buttons
- Inline button confirmations for critical actions
- Improved login design and new logout button
- File browser enhanced with rename and file actions dropdown
Git projects
- Git-based projects with clone authentication for public and private repositories
Four new LLM providers: CometAPI, Z.AI, Moonshot AI, and AWS Bedrock
Microsoft Dev Tunnels integration for secure remote access
User data migration to /usr directory for cleaner separation of user and system files
Subagents system with configurable agent profiles for different roles
Memory operations offloaded to deferred tasks for better performance
Environment variables can now configure settings via A0_SET_* prefix in .env
Automatic migration with overwrite support for .env, scheduler, knowledge, and legacy directories
Projects support extended to MCP, A2A, and external API
Workdir outside project support for more flexible file organization
Agent number tracking in backend and responses for multi-agent identification
Many bug fixes and stability improvements across the UI, MCP tools, scheduler, uploads, and WebSocket handling

v0.9.7 - Projects

Release video

Projects management
- Support for custom instructions
- Integration with memory, knowledge, files
- Project specific secrets
New Welcome screen/Dashboard
New Wait tool
Subordinate agent configuration override support
Support for multiple documents at once in document_query_tool
Improved context on interventions
Openrouter embedding support
Frontend components refactor and polishing
SSH metadata output fix
Support for windows powershell in local TTY utility
More efficient selective streaming for LLMs
UI output length limit improvements

v0.9.6 - Memory Dashboard

Release video

Memory Management Dashboard
Kali update
Python update + dual installation
Browser Use update
New login screen
LiteLLM retry on temporary errors
Github Copilot provider support

v0.9.5 - Secrets

Release video

Secrets management - agent can use credentials without seeing them
Agent can copy paste messages and files without rewriting them
LiteLLM global configuration field
Custom HTTP headers field for browser agent
Progressive web app support
Extra model params support for JSON
Short IDs for files and memories to prevent LLM errors
Tunnel component frontend rework
Fix for timezone change bug
Notifications z-index fix

v0.9.4 - Connectivity, UI

Release video

External API endpoints
Streamable HTTP MCP A0 server
A2A (Agent to Agent) protocol - server+client
New notifications system
New local terminal interface for stability
Rate limiter integration to models
Delayed memory recall
Smarter autoscrolling in UI
Action buttons in messages
Multiple API keys support
Download streaming
Tunnel URL QR code
Internal fixes and optimizations

v0.9.3 - Subordinates, memory, providers Latest

Release video

Faster startup/restart
Subordinate agents can have dedicated prompts, tools and system extensions
Streamable HTTP MCP server support
Memory loading enhanced by AI filter
Memory AI consolidation when saving memories
Auto memory system configuration in settings
LLM providers available are set by providers.yaml configuration file
Venice.ai LLM provider supported
Initial agent message for user + as example for LLM
Docker build support for local images
File browser fix

v0.9.2 - Kokoro TTS, Attachments

Release video

Kokoro text-to-speech integration
New message attachments system
Minor updates: log truncation, hyperlink targets, component examples, api cleanup

v0.9.1 - LiteLLM, UI improvements

Release video

Langchain replaced with LiteLLM
- Support for reasoning models streaming
- Support for more providers
- Openrouter set as default instead of OpenAI
UI improvements
- New message grouping system
- Communication smoother and more efficient
- Collapsible messages by type
- Code execution tool output improved
- Tables and code blocks scrollable
- More space efficient on mobile
Streamable HTTP MCP servers support
LLM API URL added to models config for Azure, local and custom providers