Learning Checkin

Daily learning habit builder with check-ins and smart reminders

Learning Check-in Skill

Help users build a daily learning habit through simple check-ins and intelligent reminders.

Overview

This skill enables users to track their daily learning with:

• Simple daily check-in (just say "I'm done" or "check-in complete")
• Automatic streak tracking
• Optional smart reminders

Data Storage

All data is stored locally in a data subfolder next to the skill:

<skill_directory>/data/
├── rule.md           - User's customizable rules
├── records.json      - Check-in history
├── version.txt       - Current skill version
├── cron_status.json  - Reminder configuration status
└── reminder_log.json - Reminder sending log

The data folder is automatically created on first use.

Commands

1. Initialize (First Time)

python <skill_path>/learning_checkin.py init

Returns:

• welcome_message - Welcome text for the user
• environment - Only contains user_language (for message display)
• reminder_strategy - Suggested reminder times
• cron_status - Current reminder configuration status

Agent action:

1. Run the init command
2. Show welcome message and explain the check-in process
3. Ask user if they want daily reminders
4. Ask user to start their first check-in

2. Check-in

python <skill_path>/learning_checkin.py checkin

Returns:

• success - Whether check-in was recorded
• streak - Current streak count
• message - Celebration message (in English, translate to user's language)

3. Status

python <skill_path>/learning_checkin.py status

Returns:

• checked_in_today - Whether user has checked in today
• streak - Current streak count
• total_checkins - Total days checked in
• message - Status message (in English)

4. Get User Language

python <skill_path>/learning_checkin.py env

Returns:

• user_language - Detected language (zh/en)

Why needed: Only to display messages in the user's preferred language.

5. Get Reminder Message

python <skill_path>/learning_checkin.py message <time>

Where <time> is one of: 09:00, 17:00, 20:00

Returns:

• message - Reminder text (in English, translate to user's language)

6. Check Reminder Status

python <skill_path>/learning_checkin.py reminder <time>

Returns:

• should_send - Whether reminder should be sent
• checked_in - Whether user has already checked in today

7. Update Cron Status

python <skill_path>/learning_checkin.py update-cron <times>

When to use: After setting up reminders (optional).

8. Get Cron Status

python <skill_path>/learning_checkin.py cron-status

Returns:

• configured - Whether reminders are set up
• times - Configured reminder times

Default Behavior

Check-in Rule

• User checks in once per day
• Simply tell the Agent "I'm done" or "check-in complete"

Reminder Strategy (Suggested)

If user wants reminders, Agent can use any scheduling method:

• Evening (20:00) is recommended as default
• Or user's preferred time

The skill will check if user already checked in before sending reminders.

Streak System

• Consecutive days = streak
• Miss a day = streak resets

Customization

Users can edit the rule.md file (in the data folder) to customize reminder messages.

Version

See GitHub releases: https://github.com/daizongyu/learning-checkin/releases

Agent Guidelines

First Interaction (Welcome)

The Agent should:

1. Be warm and encouraging
2. Explain the simple check-in process
3. Ask if user wants daily reminders (optional feature)
4. Ask: "Ready to start your first check-in?"

Check-in Interaction

• Translate messages to user's language
• Celebrate the check-in
• Show streak count

Reminder Implementation (Optional)

If user wants reminders:

• Agent decides how to implement (cron, native scheduler, etc.)
• The skill provides reminder and message commands
• Check if user already checked in before sending

Technical Notes

• Data collection: Only user_language is collected for message display
• All messages are in English - Agent translates to user's language
• All file paths use UTF-8 encoding
• Compatible with Windows, Linux, macOS
• Data stored in data subfolder next to the skill
• No external network requests from the skill
• No automatic scheduling - Agent decides implementation
• No external dependencies (Python standard library only)

Version

Current version: 3.1.0

GitHub: https://github.com/daizongyu/learning-checkin