Getting Started with AgentForge
Learn how to set up, configure, and deploy your first AI agent in under 5 minutes. This guide covers installation, core concepts, and common tasks.
On this page
Quick Start
AgentForge is a framework for building, testing, and deploying autonomous AI agents. It provides a unified API across multiple LLM providers — OpenAI, Anthropic, Google, and open-source models.
To get started, install the CLI:
npm install -g @agentforge/cli
agentforge init my-agent
cd my-agent
agentforge dev
This creates a new agent project with a default configuration, a hello-world agent, and a development server that auto-reloads on changes.
Prerequisites
- Node.js 18+ — JavaScript runtime for running agents locally
- An API key — From OpenAI, Anthropic, or any supported provider
- Git — For version control and template management
Installation
AgentForge supports multiple installation methods depending on your environment. The npm package is the recommended approach for most users.
npm (Recommended)
npm install -g @agentforge/cli
Docker
docker pull agentforge/cli:latest
docker run -it agentforge/cli init
From Source
git clone https://github.com/agentforge/agentforge.git
cd agentforge
npm install
npm run build
npm link
Configuration
AgentForge uses a YAML configuration file (agentforge.yaml) at the root of your project. The CLI generates a default config during agentforge init.
# agentforge.yaml
provider: openai
model: gpt-4o
temperature: 0.7
max_tokens: 4096
tools: [web-search, file-io, code-exec]
Agents
An agent is an AI entity configured with a system prompt, a model, and a set of tools. Agents process tasks autonomously or within workflows.
Creating an Agent
// agent.js
const agent = new Agent({
name: 'research-assistant',
systemPrompt: 'You are a research assistant.',
model: 'gpt-4o',
tools: ['web-search', 'file-io']
});
Tools
Tools are functions that agents can call during task execution. AgentForge includes built-in tools and supports custom tool creation.
Workflows
Workflows chain multiple agents and tools into a multi-step process. Each step passes its output to the next step.
Memory
Memory gives agents persistent context across sessions. AgentForge supports in-memory, file-based, and database-backed storage.
Integrations
AgentForge integrates with Slack, GitHub, Notion, Gmail, and 200+ other services through its tool integration system.
Deployment
Deploy agents as API endpoints, scheduled tasks, or event-driven services. AgentForge supports Docker, Kubernetes, and serverless deployments.
Monitoring
Built-in observability with real-time agent tracing, token usage tracking, and performance metrics. Export to your existing monitoring stack.
Security
Role-based access control, end-to-end encryption, audit logging, and SOC 2 compliance. Enterprise-grade security out of the box.
API Reference
Complete REST and WebSocket API documentation. All endpoints support JSON and are versioned for backward compatibility.
CLI Commands
The AgentForge CLI provides commands for project scaffolding, local development, testing, and deployment.
Troubleshooting
Common issues, error codes, and their solutions. If you don't find what you're looking for, open a GitHub issue.