Guide 1: First Setup

Before you can run a pipeline, you need two things configured: a driver (which AI CLI tool to use) and a skill (what to ask it to do). This guide walks through both.

Prerequisites

• fluxaOS is running and you can reach the UI
• The daemon is running (if you see pipeline runs stuck at "Pending", the daemon is down — see Daemon reference)
• You have an org, user, and project already created (these are seeded by default)

Step 1: Configure a driver

A driver tells fluxaOS how to invoke an AI CLI tool. fluxaOS ships with a seeded "Claude Code" driver. If you want to use it as-is, skip to Step 2.

To create a new driver or review the existing one:

1. Navigate to Settings → Drivers
2. Click New Driver (or click an existing driver to expand it)
3. Fill in the required fields:
   • Name: A display name (e.g., "Claude Code")
   • Slug: A unique machine-readable ID (e.g., claude-code)
   • Binary: The CLI command (e.g., claude)
   • Prompt transport: How the prompt is delivered — use stdin for Claude Code
   • Output format: Use stream-json for Claude Code
   • Context layout: JSON describing which files to write to the workspace. For Claude Code, use:

     { "instructionsFile": "CLAUDE.md", "contextFile": "context.md" }

4. Click Create (or Save if editing)
5. Make sure Is Enabled is checked — disabled drivers can't be used in pipeline stages

For a full field reference, see Drivers.

Step 2: Create a skill

A skill is the prompt template that gets sent to the driver. Start simple.

1. Navigate to Settings → Skills
2. Click New Skill
3. Fill in:
   • Name: Something descriptive (e.g., "research")
   • Scope: global (available across all projects) or project (this project only)
   • Prompt template: The prompt to send. Example for a research skill:

     Read the issue carefully. Identify the affected code areas, any relevant context,
     and what a good implementation approach would look like. Write a concise plan.
     Output your result document when done.

4. Click Create

Repeat to create additional skills (e.g., "implement", "review") as needed for your pipeline.

---

Next: Guide 2: Build a Pipeline →