AI Watch · AI Code Tutorials
Brief the agent.
Keep the judgement.
Coding agents can now run real operations on real infrastructure. The hard part is no longer typing the commands; it is knowing what to ask for, in what order, and what has to be true before the next step. Each tutorial explains one job in plain terms and ends with a prompt you can copy.
The tutorials
Start here.
Rotate Postgres and Redis passwords on Railway with an AI agent.
Brief a coding agent to change database passwords through a private tunnel, without restarting the database and with seconds of exposure instead of minutes.
- 0 Inspect, read-only
- 1 Generate secrets privately
- 2 Stage without deploying
- 3 Redeploy, switch on cue
- 4 Verify twelve checks
- 5 Clean up
How they work
Plain words, then a prompt.
Explained first
What is going on, in plain terms, with diagrams. You should understand the job before you delegate it.
Safe by construction
Read-only first, gates between phases, nothing secret printed. The agent asks before it touches anything live.
Tested on real systems
Every method has been run on production apps before it is published here.
A prompt to copy
One click gives you a brief to paste into Claude Code, Codex, Cursor or any agent that runs shell commands.
Want an agent workflow built around your own stack?
We set these up with teams in our AI workshops: the briefs, the guard rails and the habits.