WHAT AGENTS CAN DO · DAILY PLAYBOOK
Issue #019 · Nov 8, 2026 · 7 min read
My favorite one today is a dad who wanted Grok Bot to set his Tesla navigation from his schedule, even though there's still no official Tesla hookup. He had it book a dentist appointment — it texted the office from his Mac — and seven minutes later the confirmation was in. The bot put the address and time on his calendar, he got in his Cybertruck, and the stop was already on the map — I like that he used the calendar the car already reads instead of waiting for a special integration.
— Gary
START HERE
New to agents? Try #2, A Suitcase Bought on eBay in 2019 Broke: Grok Bot Found the Receipt, Ruled Out Disputes and Offered to Write the Samsonite Warranty Request. It's the easiest one today.
 
SPONSORED · Your brand here — One sponsor per daily issue.
TODAY'S TOP RESULT
7 minutes later, the stop was already on the Cybertruck map

1. 7 Minutes Later, the Dentist Was on the Map: Grok Bot Texted the Office and Used Tesla Calendar Sync, With No Official App Connection

Intermediate: a bit of setup · Source-linked: links to the original post · Travel

A dad wanted Grok Bot to set his Tesla navigation from his schedule, even though there is still no official Tesla app connection. He asked it to book a dentist appointment. It texted the office from his Mac. Seven minutes later the confirmation came in. Grok Bot read the text and added the event to his calendar with the address and time. When he got in his Cybertruck, the appointment was already on the map. He tapped Start Driving and left. His setup: turn on phone calendar write access in Grok Bot, turn on Calendar Sync in the Tesla app under Profile then Settings, and give the Tesla app Full Access to calendars on the iPhone. He says the key is that the bot always includes the address and time in the event, which is how the car picks up the destination.

Try it in Grok Bot →
 
MORE REAL RESULTS
Bought on eBay in 2019

2. A Suitcase Bought on eBay in 2019 Broke: Grok Bot Found the Receipt, Ruled Out Disputes and Offered to Write the Samsonite Warranty Request

Beginner: easy, no coding · Source-linked: links to the original post · Shopping

His suitcase broke. He had bought it on eBay in 2019. He asked Grok Bot for help, and it searched his old emails. It found the order and the receipt from 2019, said eBay and card disputes were too late, and noted that Samsonite's warranty could still work. It gave him the steps and offered to write the request, and it set a reminder for Friday to finish it. A later reply asked whether Samsonite had responded; he had not posted an update by November 8, 2026, so the claim should not be treated as filed or approved.

Try it in Grok Bot →
 
About 183 estimated hours this week, 133 tied to live output

3. About 183 Estimated Hours This Week, 133 Tied to Live Output: A Weekly Grok Bot That Ranks Which Bots Actually Save Time

Intermediate: a bit of setup · Source-linked: links to the original post · Business ops

He runs multiple Grok Bots and wanted to know which ones were actually saving time, without reading every chat. He built a bot called Bot Hours Saved that once a week reads the last seven days of bot conversations and routine runs, looks for concrete completed work, and estimates how long that work would have taken a person. Empty runs, acknowledgements and chatter do not count. Every Friday it updates one report with a ranking by estimated hours saved, the work and assumptions behind each estimate, idle or overlapping bots, suggestions to pause or merge, work waiting on approval, unresolved actions, reported revenue status, and a weekly trend chart. It separates gross hours from realized hours, so finished work still waiting to go live is not treated the same as something already shipped. This week's estimate was about 183 hours of work, including 133 hours tied to live output. He says those are estimates of equivalent human effort, not 133 measured hours added back to his calendar, and that undocumented work can be missed. The report stays on the same Page; chat gets the headline numbers and a link. It recommends changes but does not pause or change other bots itself.

Try it in Grok Bot →
 
64 emails in 10 days, 15 never touched

4. 64 Emails in 10 Days, 15 Never Touched: An n8n Inbox Assistant That Only Drafts Replies, and Why Polling Missed the Rest

Intermediate: a bit of setup · Source-linked: links to the original post · Inbox & communication

A builder removed the test-sender filter on an n8n AI email assistant and let it run on a real business inbox for 10 days. The workflow is a Gmail Trigger that polls every minute, then a Gemini text classifier, a label per branch, a Gemini draft writer and Gmail's create-draft action. Nothing is sent automatically. Result: 64 emails arrived, 49 were sorted, and 15 were never touched. Every execution showed green, so nothing flagged the misses. He traced the 15 to five causes: the trigger only reads unread mail, so anything he opened on his phone first was skipped; polling is forward-only, so mail that arrived while it was down was gone; Max Emails per Poll was set to 5; the Gemini node had no retry, so one 503 lost an email; and his Google OAuth app was still in Testing, so the Gmail credential expired after 7 days and the trigger stopped. His fixes: Retry On Fail, 3 tries, 5 seconds apart, on the classifier and the draft writer; a separate daily workflow that searches in:inbox -label:ai-processed newer_than:1d and emails him anything the bot skipped, kept separate so it still runs when the assistant is down; and moving the OAuth app to In production. He published both workflows as free JSON.

Try it in Grok Bot →
 
10 runs, 8 signed off, 7 passed the real gate

5. 10 Runs, 8 Signed Off, 7 Passed the Real Gate: Three Claude Code Seats Argue Over the Spec Before Anyone Writes Code

Advanced: technical, uses coding tools · Source-linked: links to the original post · Coding

A developer kept getting code that was well written for the wrong plan, so he built a setup where three AI seats argue over a product brief before anyone writes code: a Lead PM writes the brief, a Chief Engineer checks every claim against the real repo and has to cite a file and line for each objection, and a Chief Architect with tools turned off rules on deadlocks and reviews the agreed document. He approves the signed-off document himself; only then does an Engineer implement it in an isolated clone, and a harness runs the project's real test commands. Last week he ran ten councils in solo mode, meaning every seat was Claude Code (Lead PM on Sonnet, Chief Engineer on Opus, Chief Architect on Sonnet with tools off), with at most 4 rounds, on local copies of a toy slugify library and the tool itself. 8 of 10 signed off. 7 of those 8 passed their gate. Two stalled after the Architect raised a late concern and the revision budget ran out. One signed off and then ended BLOCKED because the project's configured npm test was still broken, even though the plan's own tests passed. The median time to a sign-off or a stall was 6.5 minutes. In all six slugify runs, the Chief Engineer's first objection was that npm test ran 0 of 2 existing tests on Node 22. In a later comment he reported a single-agent baseline on the same 10 tasks: the lone agent also caught and fixed the broken test script in all 6 playground runs, and it passed the real gate 10 of 10 times, against the council's 7 of 10. He says that on small, well-specified tasks the debate did not change outcomes, and that the fair test is harder, vaguer tasks.

Try it in Grok Bot →
 
STEAL THIS PROMPT

A Suitcase Bought on eBay in 2019 Broke: Grok Bot Found the Receipt, Ruled Out Disputes and Offered to Write the Samsonite Warranty Request

Copy this and paste it into Grok Bot:

My [item] just broke. I think I bought it from [store or site] around [year]. Search my email for the original order and receipt. Tell me whether a return, card dispute or manufacturer warranty is still possible and why, with the dates you found. If a warranty claim still looks open, give me the steps and draft the request for me to review before anything is sent. Set a reminder for [day] to finish it.
Try it in Grok Bot →
IN THE WILD
“Got in my Cybertruck and the appointment was already on the map. Tapped "Start Driving" and drove. Grok Bot handled everything, A to Z.”
NEW ON THE SITE
927 real use cases are on the site now, 37 of them new since the last issue. Browse them
TOMORROW
Five more real agent workflows, pulled from what people actually built and posted this week.
 
FREE COURSE
Grok Bot 101
Set up your first agent, step by step. Plain language, no coding.
Start the free course →
You're getting this because you subscribed at WhatAgentsCanDo.com.
What Agents Can Do, 1427 North Main St, Suite A #1053, Logan, UT 84341, United States