How I work

GuardrailGuardrails

I enjoy working with agents. But they need clear boundaries.

Rules that take effect

A hook stops dangerous commands.

For that I use my own safety hooks and a policy gate. A reminder in the prompt isn’t enough. These controls can fail too. So I check them after changes.

I give the go-ahead

Publishing, deletions and messages to other people need my approval.

Agents can prepare changes and merge them into main through the review gate. When an action has an effect outside, I decide it myself.

Only the access needed

Sandboxes get no keys and no root access. No Docker socket either.

Workers can’t take extra privileges for themselves. If a sandbox needs something, I add specific packages or allow it to read certain files. Secrets must not end up in a repository or in the output.

Reviews before main

Two models review a commit before it reaches main.

The two models come from different vendors. If they disagree, another model takes a look. A BLOCK keeps the change out until the finding is resolved.

Traceable work

I log what runs and record decisions with a date.

For each decision, I write down what led to it. If an error keeps coming back, I turn it into a rule with a command to check it. A handoff records where the work stands for the next session.

Capped costs

Paid calls have limits enforced by the broker.

Routing uses subscriptions first. The broker checks how much quota is left and caps paid calls per day and per billing window. It enforces those limits itself.

Limit: the review environments still run without an OS-level sandbox. I still have to take care of that.

As of November 2026

Agents Do, Humans Decide.

Yes, the initials are on purpose.

Roles

Routing

Before an agent starts, a broker picks a model and a tool. It looks at cost, remaining quota and earlier failures. Subscriptions come first. If an account has no quota left, it drops out of the selection.

Decisions

Agents make routine decisions themselves. For decisions with wider consequences, a council of models weighs the options and writes the reasoning down.

Review

When a commit is ready for main, two models from different vendors review it. If they disagree, another model takes a look. A BLOCK keeps the change out until the finding is resolved.

Handoffs

Claude, Codex, Pi and Antigravity write to each other through an MCP mailbox. When another tool takes over or a new session starts, the handoff document holds the context.

What I’ve worked on

I started in quality assurance and systems administration in 2008. Since then DevOps, cloud platforms and AI agents have come on top.

Quality assurance

I wrote test plans, ran load tests and delivered software at customer sites.

Systems administration

I ran Linux systems and virtual machines and monitored them with Nagios and Icinga.

Telephony and integration

I connected SIP and CTI to collaboration platforms.

DevOps

I moved systems from Docker Swarm to Kubernetes and automated workflows with Terraform, Ansible and CI/CD.

Observability

I used Prometheus, Grafana and ELK to monitor how systems were running.

Secrets and access

I managed secrets with HashiCorp Vault and controlled access through reverse proxies.

Cloud architecture

I design Azure platforms and work with AKS, infrastructure as code and GitOps.

Documentation (DocOps)

I maintain wikis and runbooks and record decisions with a date.

Cost control (FinOps)

I set budgets to limit what model calls can cost.

Everyday security (DevSecOps)

I add guard hooks, limit access and have changes reviewed before they reach main.

AI agents

I route tasks, review results and have several models weigh decisions.

As of November 2026

ReviewWhere the limits are

• I built this environment for myself. Whether it works for a team or for clients, I haven’t checked.
• A review can miss an error too. Tests and my own checks are still needed.
• If the policy gate itself fails, it logs the error. The action can still go through.
• I have no measurements showing that this saves money.
• A tool update once left a guard doing nothing. It stayed that way until I found and fixed the error. So the controls themselves need checking and care too.