Checkpoint: Verify Your Setup

Time to prove your OpenClaw instance is working.

The Test

Complete these three tasks:

Task 1: Identity Check

Start a chat and ask:

"Who are you and what's your purpose?"

Pass criteria:

• Agent responds with name from SOUL.md
• Describes its role correctly
• Doesn't give generic "I'm an AI assistant" response

Task 2: Memory Persistence

Tell your agent:

"My favorite programming language is Python and I'm working on a project called Phoenix."

End the session. Start a new one. Ask:

"What's my favorite language and current project?"

Pass criteria:

• Agent correctly recalls Python
• Agent correctly recalls Phoenix
• Information survived session restart

Task 3: File Verification

Check your workspace:

Pass criteria:

• Today's log file exists
• Contains mention of Python and/or Phoenix

Scoring

• 3/3: Your setup is solid. Continue!
• 2/3: Minor issue. Review the failing area.
• 1/3: Revisit the setup guide.
• 0/3: Start over from "Recommended Setup"

Module Summary

You've learned:

• ✅ The three setup options (local, VPS, Docker)
• ✅ How to install and initialize OpenClaw
• ✅ The directory and config structure
• ✅ Where the "brain" actually lives
• ✅ How to validate your setup

Key Insight: The LLM doesn't remember — your files do. Everything important goes in workspace/.

Next: We'll configure your models strategically.