Skip to content

Installation ​

This guide will walk you through the process of installing the Factifai Agent Suite and its dependencies.

Prerequisites ​

Before installing the Factifai Agent Suite, ensure you have the following prerequisites:

  • Node.js 18+ - The runtime environment for JavaScript
  • pnpm 10.11.0+ - The package manager used by Factifai Agent Suite
  • OpenAI API key or Azure OpenAI credentials or AWS Bedrock credentials - Required for the AI capabilities

Installing Factifai Agent ​

The Factifai Agent is the main CLI tool that you'll use to run tests. You can install it globally using npm:

bash
# Install factifai-agent globally
npm install -g @presidio-dev/factifai-agent

Installing Playwright Dependencies ​

Factifai Agent uses Playwright under the hood for browser automation. You'll need to install Playwright globally first, then install the browser dependencies:

bash
# Install Playwright globally (required)
npm install -g playwright

# Install Playwright dependencies (required)
npx playwright install --with-deps

This process installs browser binaries for Chromium, Firefox, and WebKit, along with any necessary system dependencies.

Installing Playwright Core (Optional) ​

If you want to use the Playwright Core package separately, you can install it using npm:

bash
# Install playwright-core
npm install @presidio-dev/playwright-core

Configuring API Keys ​

OpenAI ​

To use Factifai Agent with OpenAI models, you need to configure your API key:

bash
# Configure OpenAI API key (only needed once)
factifai-agent config --set OPENAI_API_KEY=your-api-key-here

Azure OpenAI ​

To use Factifai Agent with Azure OpenAI models, you need to configure your API key, instance name, deployment name and API version:

bash
# Configure Azure OpenAI API key, instance name, deployment name and API version (only needed once)
factifai-agent config --set AZURE_OPENAI_API_KEY=your-api-key-here
factifai-agent config --set AZURE_OPENAI_API_INSTANCE_NAME=your-instance-name
factifai-agent config --set AZURE_OPENAI_API_DEPLOYMENT_NAME=your-deployment-name
factifai-agent config --set AZURE_OPENAI_API_VERSION=your-api-version

AWS Bedrock ​

To use Factifai Agent with AWS Bedrock models, you need to configure your AWS credentials:

bash
# Configure AWS credentials (only needed once)
factifai-agent config --set AWS_ACCESS_KEY_ID=your-access-key-id
factifai-agent config --set AWS_SECRET_ACCESS_KEY=your-secret-access-key
factifai-agent config --set AWS_DEFAULT_REGION=us-west-2

Configuring Secrets ​

In addition to API keys, you may want to store other sensitive information like test account credentials or API tokens that will be used during testing. Factifai Agent provides a dedicated secret management system for this purpose:

bash
# Store a test account username
factifai-agent secret --set TEST_USERNAME=test_user@example.com

# Store a test account password
factifai-agent secret --set TEST_PASSWORD=secure_password123

# Store an API token for testing
factifai-agent secret --set TEST_API_TOKEN=your-api-token-here

These secrets are stored separately from configuration in ~/.factifai/secret.json and are automatically loaded as environment variables during test execution.

To view all stored secrets (with values masked):

bash
factifai-agent secret --list

To delete a secret when it's no longer needed:

bash
factifai-agent secret --delete TEST_PASSWORD

Development Setup ​

If you want to contribute to the Factifai Agent Suite or run it from source, follow these steps:

bash
# Clone the repository
git clone https://github.com/presidio-oss/factifai-agent-suite.git
cd factifai-agent-suite

# Install dependencies
pnpm i

# Build all packages
pnpm -r build

# Install Playwright dependencies (required)
npx playwright install --with-deps

# Create a global symlink for factifai-agent
cd packages/factifai-agent
pnpm link --global

Verifying Installation ​

To verify that the installation was successful, run:

bash
factifai-agent --version

This should display the version of the Factifai Agent that you have installed.

Next Steps ​

Now that you have installed the Factifai Agent Suite, you can proceed to the Quick Start Guide to run your first test.

Released under the MIT License.