OpenSidebar
OpenSidebar Ideas

Ideas for agents
that deserve trust.

Essays on reliability, human agency, evaluation, and the systems that turn capable models into dependable software.

01 published 01 in progress
Writing

The essays

Long-form arguments, grounded in the product and its code.

01 Published Reliability

Done Means Verified

An agent is not finished because it acted. It is finished when it has evidence that the user’s objective was achieved.

By Kris Shkodrani & Sol gpt-5.6-sol 10 minute read
Read the essay
The reliability loop
01
ActChange the world
02
ObserveRead the new state
03
VerifyEarn the verdict
02 In progress Agency

Autonomy Needs Boundaries

A trustworthy agent is not one that constantly asks permission. It is one that can work independently inside a mandate—and recognize the moment when acting would create a consequence the user should control.

The next essay
Editorial lens

What we examine

Not model capability in isolation, but the system around it.

01 / Evidence

What makes a result believable?

Verification, read-backs, traces, and the difference between an action taken and an outcome achieved.

02 / Agency

Where should autonomy end?

Mandates, reversibility, approval boundaries, and preserving the consequential decisions for people.

03 / Systems

What turns a model into a product?

Harnesses, tools, recovery, evaluation, and the architecture required for dependable work on the web.