Skip to content

Getting Started

This guide walks you through setting up SWAO and running your first workload assessment.

New to SWAO? Visit the SWAO product page for a business-focused overview, feature comparison by edition, and stakeholder benefit summaries.

SWAO Architecture

SWAO Architecture Overview

SWAO Workflow

mermaid
flowchart LR
    A["Input\nSource Code\nConfig Files"] --> B["SWAO\nCLI · TUI · MCP"]
    B --> C{"Compliance\nFramework"}
    C --> D[GDPR]
    C --> E[HIPAA]
    C --> F[ISO 27001]
    C --> G[...]
    B --> H["Output\nHTML Report\nBI Export · JSON"]
    H --> I["7R Migration\nRecommendations"]
    I --> J["Onboarding\nTerraform · LZ\nMeshcloud"]
    J -.->|"Continuous\nImprovement"| A

Prerequisites

Before you begin, ensure you have the following installed:

  • Node.js 20 or later -- download from nodejs.org
  • pnpm -- install with npm install -g pnpm
  • An LLM API key -- SWAO uses an LLM provider for AI-assisted analysis. Supported providers include OpenAI, Azure OpenAI, Ollama (local), and Anthropic.
  • Playwright (optional) -- required for the web-crawler ingestion path. Install with npx playwright install.

Installation

Binary downloads for SWAO Community Edition are coming in Phase 2 of the release programme. Check the GitHub Releases page for the latest availability updates.

In the meantime, you can run SWAO directly from source:

bash
git clone https://github.com/Accenture/SWAO.git
cd SWAO
pnpm install
pnpm build

Workspace Initialisation

Before running your first assessment, initialise a SWAO workspace in your project directory. This is a one-time setup step that creates the configuration file and wires up your LLM provider.

bash
cd my-project
swao init

The swao init wizard prompts you for:

  1. App name -- a short identifier for this application (used in reports and the TUI header)
  2. Compliance framework -- the framework to assess against (e.g. gdpr, hipaa, iso27001)
  3. Source path -- path to the application source code relative to the workspace root
  4. LLM provider -- choose from openai, azure-openai, anthropic, ollama
  5. LLM API key -- pasted interactively and stored in a local .env file (never committed)
  6. Playwright -- whether to enable the web-crawler ingestion path

On completion, swao init creates:

  • .swao.yml -- workspace configuration
  • .env -- LLM credentials (add to .gitignore before committing)
  • .gitignore entry for .env and swao-output/

Example .swao.yml after init

yaml
app: payment-service
framework: gdpr
sourcePath: ./src
outputPath: ./swao-output

llm:
  provider: anthropic
  model: claude-sonnet-4-6

playwright:
  enabled: false

Quick Start

Follow these five steps to complete your first assessment:

Step 1 -- Initialise your workspace

Run the init wizard to create your workspace configuration and connect your LLM provider:

bash
mkdir my-assessment && cd my-assessment
swao init

Follow the interactive prompts. See Workspace Initialisation above for a full walkthrough of each prompt.

Step 2 -- Verify your environment

Run the diagnostics command to confirm that SWAO can reach all required dependencies:

bash
swao doctor

Fix any issues flagged before proceeding.

Step 3 -- Run an assessment

Execute a compliance assessment against your configured application:

bash
swao assess --app <name>

Replace <name> with the application identifier defined in your .swao.yml. SWAO will analyse the source path and emit findings for each applicable control.

Step 4 -- Open the HTML report

After the assessment completes, open the generated HTML report in your browser:

bash
swao report --open

The report contains every finding, its evidence link, and a summary by control domain.

Step 5 -- Explore the interactive TUI

For a live, navigable view of findings without leaving the terminal, run the assessment in interactive mode:

bash
swao assess --interactive

Use the arrow keys to navigate findings and press q to quit.

Next Steps

  • CLI Reference -- full command and flag documentation
  • TUI -- explore findings interactively without leaving the terminal
  • MCP Server -- connect SWAO to Claude Code or any MCP-compatible AI client
  • HTML Report -- understand the audit-ready HTML output
  • Power BI Export -- build compliance dashboards in Power BI
  • Configuration -- all .swao.yml options explained
  • Frameworks -- supported compliance frameworks

Migration pathways (7R)

mermaid
quadrantChart
    title 7R Cloud Migration Strategies
    x-axis Low Effort --> High Effort
    y-axis Low Business Value --> High Business Value
    quadrant-1 Transform
    quadrant-2 Invest
    quadrant-3 Retire
    quadrant-4 Lift and Shift
    Retire: [0.1, 0.1]
    Retain: [0.2, 0.5]
    Rehost: [0.3, 0.35]
    Relocate: [0.4, 0.45]
    Replatform: [0.6, 0.65]
    Refactor: [0.85, 0.85]
    Repurchase: [0.55, 0.55]

SWAO is an Accenture open-source project.