Session 2 · Automate with Governance
Session 2 · Automate with Governance
Info
Tip
Week 2 of 4. Last week you standardized your team with skills and plugins. Today you scale the next link: automate with governance. You’ll stop running every task yourself and put Cowork to work on a cadence —but with clear rules about who approves what.
The CEO no longer wants reports by hand: he wants your division’s AI operation to run on its own, under control. You’ll schedule a weekly Elektra Rewards KPI report, define your team’s permission-mode policy, and meet Dispatch (control from your phone) and native sub-agent coordination.
By the end of the session you’ll have a live scheduled task, a written mode policy (what goes in Manual, what in Auto, nothing in Skip without approval), and the judgment to govern automation without losing control.
Agenda
| Block | Topic | Duration |
|---|---|---|
| 1 | Scheduled tasks in depth: build your weekly report | ~35 min |
| 2 | Modes and governance: the Manual/Auto/Skip matrix | ~45 min |
| 3 | Dispatch: 🎥 Instructor demo (Pro/Max only) | ~15 min |
| 4 | Sub-agents: native coordination (informational) | ~10 min |
| 5 | Wrap-up: checkpoint, quiz, and bridge | ~15 min |
| Core | ~120 min | |
| ⏱ optional | If there’s time: the limits of memory | ~15 min |
Learning objectives
- Create and manage a scheduled task (Create with Claude and manual setup), understanding what runs remotely vs. locally.
- Read the Manual / Auto / Skip matrix against each connector’s permission and draft your team’s mode policy.
- Explain what Dispatch is, its plan limits (Pro/Max), and the phone→desktop safety chain.
- Recognize that sub-agents are native coordination Claude decides on, not something you launch.