Founder
Apple Reminders Set From Telegram β "Remind Me Friday" and It Just Works
Key Takeaway
One Telegram message to my AI agent creates an Apple Reminder or Things 3 task β synced across every device I own β in 5 seconds flat.
The Problem
I live in Telegram. My task management lives in Apple Reminders and Things 3. That's two different apps, two different mental contexts, and a friction tax I pay every single time I need to remember something.
Here's what "just setting a reminder" actually looks like without an agent:
- Leave Telegram. Open Reminders.
- Tap the plus button.
- Type the title.
- Tap the date field. Scroll to Friday.
- Tap the time field. Set 10:00 AM.
- Save. Switch back to Telegram.
One minute. Minimum. And that's if I don't get distracted by whatever else is on my screen. Multiply by 15 tasks a day and you've burned 15 minutes doing data entry that adds zero value.
Things 3 is worse. Project assignment, tags, deadlines β each one is another tap. The app is beautiful. Using it as an input mechanism is not.
The real cost isn't the time. It's the context switch. I'm mid-conversation, a task surfaces, and instead of capturing it in flow, I have to break state. Half the time I tell myself "I'll add it later." Later never comes.
The Solution
I message Alfrawd in Telegram. Natural language. Done.
Alfrawd is my master agent running on Mr.Chief. He sits on a Linux gateway, but he has a paired node β my Mac. That Mac node can talk to Apple Reminders, Things 3, Notes, anything local. The architecture is simple: Telegram β agent β Mac node β Apple app β syncs everywhere.
No app switching. No date pickers. No friction.
The Process
The setup is a Mac node paired to the Mr.Chief gateway via Tailscale. The agent has allowlisted commands on the Mac, including osascript for AppleScript execution and the Things 3 URL scheme.
When I send a message like:
View details
remind me to call Christophe about Stealth AM on Friday at 10am
Alfrawd parses the intent, extracts the structured data, and routes to the Mac node:
bashShow code
# Agent executes on Mac node (100.75.113.94)
osascript -e '
tell application "Reminders"
tell list "Reminders"
set newReminder to make new reminder with properties {
name: "Call Christophe about Stealth AM",
due date: date "Friday, March 13, 2026 10:00:00 AM"
}
end tell
end tell'
For Things 3, the pattern is similar but uses the URL scheme:
View details
add task: review Bookplayer contract, due Monday, project: Legal
Becomes:
bashShow code
open "things:///add?title=Review%20Bookplayer%20contract&when=2026-03-16&list=Legal"
The agent confirms back in Telegram:
View details
β
Reminder set: "Call Christophe about Stealth AM"
π
Friday, March 13 at 10:00 AM
π± Syncing to all devices
iCloud handles the rest. iPhone, Mac, Apple Watch β they all see it within seconds.
The Results
| Metric | Before (Manual) | After (Agent) |
|---|---|---|
| Time to create task | 45-90 sec | 5 sec |
| Context switches | 2 (leave app, return) | 0 |
| Tasks captured vs. intended | ~60% (rest forgotten) | ~98% |
| Daily time on task entry | ~15 min | ~1 min |
| Devices synced | Same (iCloud) | Same (iCloud) |
| Works mid-conversation | No | Yes |
The capture rate is the number that matters. I went from forgetting 4 out of 10 tasks to forgetting almost none. Not because I got more disciplined β because the friction disappeared.
Try It Yourself
- Set up an Mr.Chief gateway with a Mac node paired via Tailscale
- Allowlist
osascriptandopencommands on the Mac node - Send natural language task messages to your agent via Telegram
- Agent parses intent β routes to Mac β creates reminder/task β confirms
The entire setup takes 20 minutes. The daily time savings compound forever.
Your phone is a capture device, not an input form. Treat it that way.
Related case studies
Founder
ClawHub: From 15 Skills to 52 in One Afternoon β The Skill Marketplace That Scales Your Agent
Started with 15 bundled skills. ClawHub marketplace got us to 52 in one afternoon. Finance, legal, security, research β here's how we evaluated and installed 37 skills.
Product Manager
Monitoring 100 Competitor Pages for Changes β Weekly Diff Report
An AI agent scrapes 100 competitor pages weekly, diffs them against the previous snapshot, and flags changes. Pricing shifts, new features, team hires β nothing slips through.
CTO
Security Audit That Runs Every Morning β 149 Intrusion Attempts Caught on Day One
An AI agent runs a full security audit every morning at 7am β UFW, SSH, fail2ban, open ports. Day one: 149 blocked brute-force attempts. Here's how we set it up with Mr.Chief.
Want results like these?
Start free with your own AI team. No credit card required.