AI DevKit Pi Integration

AI DevKit can detect Pi sessions through its agent management system. For the most accurate integration, launch Pi at least once and run the AI DevKit machine setup:

ai-devkit setup

With npx only:

npx ai-devkit@latest setup

Setup installs the dedicated Pi session tracker and built-in AI DevKit skills for detected agents. For the complete setup-then-init sequence, see Getting Started.

@ai-devkit/pi-session-tracker helps AI DevKit identify active Pi sessions with better session metadata than process detection alone.

This matters when you use commands such as:

ai-devkit agent list
ai-devkit agent detail --id <pi-session>
ai-devkit agent send "What are you working on?" --id <pi-session>

Without the tracker, AI DevKit may still detect Pi through local processes, but the result can be less precise.

Install it if you use Pi with AI DevKit agent management features, especially if you want AI DevKit to more reliably list, inspect, or message your active Pi sessions.

If the Pi tracker step fails during setup, retry its underlying installation command:

pi install npm:@ai-devkit/pi-session-tracker

After installing, start or restart your Pi session, then check what AI DevKit can see:

ai-devkit agent list

No. Machine setup improves Pi session visibility and installs global skills. You should still initialize AI DevKit in each project when you want workflow docs and other project configuration:

ai-devkit init

• Pi package: https://pi.dev/packages/@ai-devkit/pi-session-tracker
• AI DevKit Agent Management: https://ai-devkit.com/docs/8-agent-management