Mind Window · Agent Design Studio

Mind Window

Designing how agents think — and how humans watch them think.

Most agent demos show you the output. This is a study in the other layer: the mind (how an agent decides, recalls, and refuses) and the window (how a person watches that decision happen and chooses to trust it). Each entry is one agent, built end-to-end, with its real traces, its design rationale, and an honest account of where it broke.

The Agents

One agent, built to the floor — not a template gallery

AGENT 01
The Article Analyst

An agent that reads what you read, builds a memory of it, and answers questions against the whole corpus — citing the specific source behind every claim, and flagging the claims it can't ground.

The moment that matters: mid-run, the agent detects its own search drifting off-topic, issues a RETRY, and recovers — on a real trace. Resilience, captured live, is more convincing than a demo that never stumbles.
Read the case study & explore the trace
LangGraph · 3 graphs 71-article corpus FK-grounded citations 28 sessions

Deep Dives

Seven cross-sections of how the Article Analyst was designed

Why this exists

The bet behind the studio