Skip to main content

Installation

There are two ways to set up Scraut. Both produce identical results — choose the one that fits your workflow.

Option A — npx create-scraut (recommended)​

The interactive Node.js wizard handles everything: it clones the Scraut repository, asks 9 questions about your team, and scaffolds all configuration and workspace template files in one step.

Quickstart​

npx create-scraut my-team
cd my-team

That's it. The wizard clones github.com/aldhosutra/scraut into ./my-team/, then immediately starts the interactive prompts.

Running from an empty directory

If you prefer to clone yourself first:

mkdir my-team && cd my-team
npx create-scraut

When the current directory is empty, the wizard clones Scraut into it automatically.

Already inside a cloned repo

If you ran git clone https://github.com/aldhosutra/scraut.git my-team && cd my-team manually, you can run the wizard from inside:

node apps/create-scraut/bin/create-scraut.js

The wizard detects it's already inside a Scraut repository and skips the clone step.

The wizard prompts​

  Scraut Setup Wizard

  GitHub repository (org/repo): myorg/my-repo
  Team member GitHub logins (comma-separated): alice,bob,charlie
  Product owner login: alice
  Scrum master login: bob
  Slack channel: #scraut-bot
  Sprint length:
  ❯ 2 weeks (14 days)
    1 week  (7 days)
    3 weeks (21 days)
  Starting sprint number (1 for a brand-new team): 1
  Timezone (IANA format, e.g. UTC, Asia/Jakarta): Asia/Jakarta
  LLM provider:
  ❯ anthropic
    openai
    gemini
    ollama
  Slack webhook URL (Enter to skip): https://hooks.slack.com/...

What the wizard creates​

After answering, you'll have:

my-team/
├── .github/workflows/          ← 28 automation workflows
├── apps/
│   ├── automation/             ← Python automation layer
│   ├── create-scraut/          ← npm wizard
│   └── portal/                 ← visibility portal
├── workspace/
│   ├── scraut.yml              ← your configuration
│   ├── sprint/001/
│   │   ├── standup/TODAY/
│   │   │   ├── alice.md        ← standup template for Alice
│   │   │   ├── bob.md
│   │   │   └── charlie.md
│   │   ├── retrospective/
│   │   │   ├── alice.md        ← retro template for Alice
│   │   │   ├── bob.md
│   │   │   └── charlie.md
│   │   ├── meta.md             ← sprint metadata
│   │   └── grooming/backlog-ideas.md
│   ├── team/capacity.md        ← team availability template
│   ├── okr/okr.md              ← OKR template
│   ├── customer/feedback.md    ← customer feedback log
│   └── milestones/README.md    ← milestone format guide
└── .scraut/                    ← bot-generated output (empty until workflows run)

Option B — scraut init (Python CLI)​

If you prefer Python, clone the repo manually and run the Python wizard:

Step 1: Clone​

git clone https://github.com/aldhosutra/scraut.git my-team
cd my-team

Step 2: Install Python dependencies​

pip install -r apps/automation/requirements.txt

Step 3: Run the wizard​

scraut init

The prompts are identical to the npm wizard. Same output, same template files.

After setup​

Regardless of which option you used, you now have:

  • workspace/scraut.yml — your configuration file
  • Template workspace files so every team member sees the expected format
  • .github/workflows/ — 28 automation workflows ready to activate

Next steps:

  1. Configure scraut.yml — fill in slack_id and email for each team member
  2. Set up GitHub Secrets and Pages — required for workflows to run
  3. Launch Sprint 1 — kick off your first sprint

Upgrading Scraut​

Scraut is a cloned repository, so upgrading is a git pull:

git remote add upstream https://github.com/aldhosutra/scraut.git
git fetch upstream
git merge upstream/main

Your workspace/ files are never touched by Scraut automation, so they survive upgrades safely.