Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.joggr.ai/llms.txt

Use this file to discover all available pages before exploring further.

jog init analyzes your repository and sets up everything AI tools need to work reliably — instruction files, coding standards, AI rules, architecture docs, guardrails, and an audit of your existing config.
jog init is safe to re-run at any time. The agent updates existing files rather than overwriting them, so you can re-run it after adding new dependencies, restructuring your codebase, or onboarding new AI tools.

​
Prerequisites

Complete Authenticate before continuing.

​
1. Run init

From your repository root:
Run jog init and the CLI walks you through selecting your AI tools and preferences.
jog init
The agent analyzes your codebase, asks about your team’s preferences, then generates instruction files, coding standards, AI rules, architecture docs, and agent hooks based on what it found.

​
2. Review the output

After running, you’ll see a summary like: 
Actions:
  ✓ Created CLAUDE.md
  ✓ Created .cursorrules

Recommendations:
  ⚠ [medium] Add a .editorconfig for consistent formatting across editors
  ⚠ [low] Consider adding a CONTRIBUTING.md for onboarding

Summary: 2 actions, 2 recommendations
A detailed report is saved locally for reference.

​
3. Commit the generated files

The generated files belong in version control so your teammates and AI tools can use them. 
git add .
git commit -m "chore(repo): initialize AI tooling with jog init"

​
Options

​
--providers, -p
string
Comma-separated list of AI tools: claude, cursor (more coming soon).
​
--guidance, -g
string
Custom guidance to steer the setup agent. Example:
jog init --providers claude --guidance "Focus on TypeScript best practices and monorepo patterns"

​
Troubleshooting

jog init requires authentication. Run jog auth login first. See the authentication guide.
Make sure you’re running jog init from your repository root (where your package.json or project config lives). The agent needs to see your codebase to generate useful configuration.

​
Next Step

Your repo is ready for AI-assisted development. Set up Workflows to plan and build features with a structured workflow.