| --- |
| title: Email Triage AI |
| emoji: 📧 |
| colorFrom: blue |
| colorTo: green |
| sdk: docker |
| app_file: app.py |
| pinned: false |
| --- |
| |
| # Email Triage AI Environment |
|
|
| ## Problem |
| Email overload is a real-world problem. This environment trains an AI agent to classify emails and generate replies. |
|
|
| ## Observation |
| - Email text |
|
|
| ## Action |
| - label (important / spam / normal) |
| - reply (text) |
|
|
| ## Reward |
| - Correct label: +0.5 |
| - Correct reply: +0.5 |
|
|
| ## API Endpoints |
|
|
| ### POST /reset |
| Returns a new email. |
|
|
| ### POST /step |
| Takes action and returns reward. |
|
|
| ## Description |
| This project simulates a real-world email automation system using a reward-based AI environment. |
