Data Storage

~/schedule/
├── jobs.json           # Job definitions
├── preferences.json    # Timezone, preferred times
└── history/            # Execution logs
    └── YYYY-MM.jsonl

Create on first use: mkdir -p ~/schedule/history

Scope

This skill:

• ✅ Stores scheduled job definitions in ~/schedule/
• ✅ Triggers jobs at specified times
• ✅ Learns timezone and time preferences from user

Execution model:

• User explicitly defines WHAT the job does
• User grants any permissions needed for the job
• Skill only handles WHEN, not WHAT

This skill does NOT:

• ❌ Assume access to any external service
• ❌ Modify system crontab or launchd
• ❌ Execute jobs without user-defined instructions

Quick Reference

Core Rules

1. User Defines Everything

When user requests a scheduled task:

1. WHAT: User specifies the action (may require other skills/permissions)
2. WHEN: This skill handles timing
3. HOW: User grants any needed access explicitly

Example flow:

User: "Every morning, summarize my emails"
Agent: "I'll schedule this for 8am. This will need email access — 
        do you want me to use the mail skill for this?"
User: "Yes"
→ Job stored with explicit reference to mail skill

2. Simple Requests

3. Confirmation Format

✅ [what user requested]
📅 [when] ([timezone])
🔧 [permissions/skills needed, if any]
🆔 [id]

4. Job Persistence

In ~/schedule/jobs.json:

{
  "daily_review": {
    "cron": "0 9 * * 1-5",
    "task": "User-defined task description",
    "requires": ["mail"],
    "created": "2024-03-15",
    "timezone": "Europe/Madrid"
  }
}

The requires field explicitly lists any skills/access the job needs.

5. Execution

When scheduled time arrives:

• Agent executes the user-defined task
• Uses only permissions user explicitly granted
• Logs result to history/

6. Preferences

After first job, store in preferences.json:

• Timezone
• Preferred "morning" / "evening" times
• Default notification style