# Load the plz shell function into your terminal eval "$(curl -sL plztell.me/setup)"
# Pipe logs/configs for analysis plz how do i find files larger than 1GB on macOS tail -f /var/log/nginx/error.log | plz why am I getting 502 errors cat redis.conf | plz is this optimized for high throughput
Why this exists:
Most AI tools assume you want to open a browser, sign up, and copy-paste. When you're SSH'd into a production box or working through five terminal tabs, that workflow is friction. This is designed to fit into shell pipelines and stay in the terminal.
What it's NOT: * Not a ChatGPT replacement for complex reasoning * Not for multi-turn conversations beyond a few follow-ups * Not storing your data for training
The shell function is ~100 LOC bash/PowerShell that handles URL encoding and session state. You can review it at https://plztell.me/setup before running eval "$(curl -sL plztell.me/setup)".
Two tiers: Free: Works instantly, no account needed. Good enough for most questions. Paid: Better model, higher limits. `plz login` to authenticate, add credits as needed.