VM0VM0 Documentation
Build agents and automate workflows with natural language
Welcome to VM0 documentation. Get started with building AI agents.
Quick Start
From zero to agent in 5 minutes
Product Philosophy
The vision, values, and principles that guide VM0's development
Tutorial
Overview
Create an AI agent that conducts comprehensive research and syncs reports to GitHub
My First Agent
Set up and run your first agent in 5 minutes
Writing Instructions
Define your deep research agent's workflow with custom instructions
Adding Research Capabilities
Enhance your agent with Firecrawl and Tavily for better research
Syncing Reports to GitHub
Enable your agent to automatically commit research reports to GitHub repositories
Core Concepts
Agent Anatomy
Understanding what makes up a VM0 agent
Instructions
Define workflows with natural language
Skills
Add SaaS integrations to your agent
Volume
Underlying implementation of skills and instructions
Artifact
Persistent storage for agent files
Environment Variables
Configure secrets and variables for VM0 agents
Usage
Model Selection
Overview
Configure different model providers for your agents
Claude
Use Claude models with full control over Opus, Sonnet, and Haiku selection
OpenRouter
Use multiple model providers through OpenRouter
Moonshot (Kimi)
Use Kimi K2 model with VM0
MiniMax
Use MiniMax M2.1 model with VM0
DeepSeek
Use DeepSeek models with VM0
Z.AI (GLM)
Use GLM models with VM0
Agent Skills
GitHub
Interact with GitHub repositories, issues, and pull requests
Slack
Send messages and interact with Slack workspaces
Discord
Send messages and manage Discord servers
Notion
Create and manage Notion pages and databases
Linear
Manage issues and projects in Linear
Jira
Project management and issue tracking
Supabase
PostgreSQL database with REST API
Firecrawl
Web scraping and crawling with Firecrawl API