← All playbooks
ResearchWe run this

A daily scout for new agent use cases

Each weekday, one new agent use case with traction: the problem, the tools, the steps and the sources.

For
Agencies and builders deciding what to build next
Setup time
15 minutes
Runs on
Grok Bot

Why we run it

This is how this directory grows. Grok proposes one candidate a day; we only publish it after running it ourselves.

Steps

  1. Schedule a weekday routine for 9 am.
  2. Search X and Reddit for agent setups people are actually using, not demos.
  3. Pick the one with the most real-world signal and write it up in a fixed format.
  4. Mark it 'to test'. It only becomes a playbook after you run it.
The promptIn the kit
Every weekday at 9 am, search X and Reddit for AI agent setups people say they
use in real work (not demos or announcements). Pick ONE with the strongest signal.
Write: title, problem it solves, tools, setup steps, what could go wrong, source links.
Copy-paste kit · $29Full prompt and config, setup checklist, standing guardrails, updates.

Also in The Operator's Autopilot, The Full Bench.

Guardrails

  • Nothing goes into the directory as 'we run this' until we've run it.

What broke

  • Nothing worth reporting yet. When it breaks, it goes here.

Our stack

Grok Bot scheduled routine.

Want it running on your accounts?We install and adapt it for you, from $500.
Have us install it
Next playbookReal-estate listing site that captures and routes leads →