Pinned Tweet
Tools I Personally Love:
🛸 Proton VPN (60% off) - go.getproton.me/SH1dk
My calm shield online
🔑 Proton Pass (50% off) - go.getproton.me/SH1dj
My password vault
🦑 GitKraken Pro (50% off) - gitkraken.cello.so/an1FhNXzY…
My Git sidekick
💖 You save, I create more!
I dressed up as a nun to confess the 7 deadly sins of AI agents 😈
Sloth: two agents waiting for each other. Forever.
Greed: one bill, zero idea which agent spent it.
God forgives. Logs remember everything.
youtu.be/8Q8MznholaA
#AIAgents #Halloween
Hired 3 AI deckhands: Claude Code, Codex, Copilot.
One branch. Instant mutiny 🏴☠️
New video: I take the helm with Kepler by @GitKraken. Own worktree per agent, a map of the whole fleet, and I finally see who’s asking for permission.
youtu.be/1kIKHLUsG4Y

ALT GitKraken Kepler tutorial: “Stop babysitting AI agents”. A developer in a pirate captain hat stands in front of a ship at sunset, next to a glowing green kraken holding the Kepler by GitKraken logo. The video shows how to run Claude Code, Codex and Copilot AI coding agents in parallel with git worktrees in Kepler, an agentic development environment (ADE).
An AI agent wiped a production database and every backup in seconds.
It wasn't hacked. It was trying to help, and it found someone else's token in a file.
3 controls from the AWS Agentic AI Lens that would have stopped it 👇
youtu.be/Tw3jK98AdW4
#AIAgents #AWS
In July, AWS fixed two read-only guardrails for AI agents in the same week.
One skipped its deny list after a failed startup load (CVE-2026-16584). The other hid write tools but still ran them for anyone who knew the name.
IAM held both times.
On AWS, read-only policies read more than their names suggest.
ReadOnlyAccess opens S3 objects and DynamoDB items. Even ViewOnlyAccess returns Lambda environment variables.
And CloudTrail records no object writes unless you set it up before the run.
What to set up before an AI agent gets an AWS role:
builder.aws.com/content/3KD1…
Poll, AWS Builder ID: which credentials does your AI agent use?
builder.aws.com/poll/3KD0tMX…
#AWSCommunityBuilder #AIAgents
Every agent project I built this year ended up with pgvector — usually right when someone realized retrieval was the problem, not the model.
Postgres Summit US is Sept 30–Oct 2 in NYC. The committers, not vendors. $400 for three days, Meetup26 takes $50 off.
2026.postgressummit.us/
New badge on the shelf:
IBM Advocate 2026, through the IBM Champions program.
This one came from the actual work.
Nice when that gets noticed.
Thank you to the IBM Champions team.
More coming. 💙
credly.com/badges/69ab3598-c…
Three AI agents are editing your codebase in parallel right now.
Your terminal shows you one at a time.
That gap is the end of a very old argument.
“Just use tmux,” someone says.
Sure - now five terminals stream five logs at once. But a log tells you what an agent is saying, not where it is in your code or whether it’s about to collide with the other two.
A multiplexer scales the output. It doesn’t scale your grasp of the state.
The terminal was built for one human doing one thing at a time. Agents don’t work that way.
Seeing every branch and commit on one surface stopped being a convenience and became the only way to stay in control.
I wrote about it, with GitKraken as the clearest example:
devops.pink/the-terminal-was…
#DevOps #Git #AIAgents





