Installation

Install AgentForge and set up your development environment.

Requirements

Node.js 22+ (for TypeScript)
Python 3.10+ (for Python SDK)
Redis (optional — for persistent state and queues)
PostgreSQL + pgvector (optional — for long-term memory)

Install

npm install @ahzan-agentforge/core

pip install agentforge

LLM Provider Setup

AgentForge supports multiple LLM providers. Set the API key for your provider:

export ANTHROPIC_API_KEY=your-key-here

export OPENAI_API_KEY=your-key-here

export GOOGLE_AI_API_KEY=your-key-here

No API key needed — Ollama runs locally.

# Install Ollama, then pull a model
ollama pull llama3

Optional Dependencies

For persistent state (Redis):

# Start Redis locally
docker run -d -p 6379:6379 redis

For long-term memory (PostgreSQL + pgvector):

# Start PostgreSQL with pgvector
docker run -d -p 5432:5432 -e POSTGRES_PASSWORD=postgres pgvector/pgvector:pg16

Scaffold a Project

Use the CLI to scaffold a new AgentForge project:

npx @ahzan-agentforge/core init my-agent
cd my-agent
npm install

This creates a project with a basic agent configuration, a sample tool, and a development script.

Verify Installation

import { createLLM } from '@ahzan-agentforge/core';

const llm = createLLM({ provider: 'anthropic', model: 'claude-sonnet-4-20250514' });
const response = await llm.chat({
  system: 'You are a helpful assistant.',
  messages: [{ role: 'user', content: 'Hello!' }],
});
console.log(response.content); // "Hello! How can I help you today?"

from agentforge import create_llm

llm = create_llm(provider="anthropic", model="claude-sonnet-4-20250514")
response = llm.chat(
    system="You are a helpful assistant.",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.content)  # "Hello! How can I help you today?"

AI Coding Skill (Optional)

If you use an AI coding assistant (Claude Code, GitHub Copilot, Gemini CLI), install the AgentForge skill so it can write correct code for you:

npx skills add ahzan-dev/ahzan-agentforge@agentforge -g -y

This teaches your assistant the full AgentForge API — tool definitions, agent config, providers, testing, multi-agent patterns, and more.

Next Steps

Quickstart — build your first agent
Project Structure — understand the file layout
Configuration — customize AgentForge settings

Installation

On this page