Set up the harness

Bring SaaSFoundryAI's development workflow to a repository you already own. Your application code stays in place: the harness profile adds shared agent instructions, skills and a managed project manifest, not a replacement API or frontend.

1. Prepare your machine

Install Git, Node.js 24.19.0 and npm 11. You do not need Docker just to install the harness. A coding-agent runtime is optional if you will use the CLI yourself.

2. Install in your repository

cd your-project
npx saasfoundryai-cli new --profile harness

The interactive setup asks which coding-agent profiles and team tools to configure. Review its recap before confirming. It writes .saasfoundry.json and reviewable project instructions; it does not replace your technical stack.

Starting a new SaaS instead? Run npx saasfoundryai-cli new and choose full to combine the harness with the ready SaaS foundation, or stack for the foundation without the managed workflow.

3. Verify

npx saasfoundryai-cli status --agent-friendly --no-network
npx saasfoundryai-cli agents list --json
npx saasfoundryai-cli docs

The last command opens the documentation bundled with the CLI, even without the hosted site. If you prefer the short sf command, install once with npm install -g saasfoundryai-cli.

Prefer to ask an assistant?

In Claude Code, open the repository and say:

Install the SaaSFoundryAI skill from https://github.com/DiamondForgeFr/SaasFoundryAI, then help me set up the development harness in this repository.

The assistant proposes the setup and asks for approval before running the same CLI. This one-line skill bootstrap is currently Claude Code-specific. With Codex, Gemini CLI, Kimi Code, Qwen Code or another coding agent, run the terminal steps above, select that agent's profile, then open the configured repository in the agent.

Need the decision matrix, platform notes or a new-project walkthrough? Continue to the full installation guide or CLI versus assistant setup.