MARGIN

MARGIN began with a product-design question: what should a web tool do when a person has very little attention, very little patience, and one difficult decision directly in front of them?

Most digital health products are organized around accounts, assessments, programs, dashboards, streaks, or long-term records. Those structures can be useful in the right setting. They can also become one more demand during a high-pressure moment.

MARGIN takes a narrower position. It does not ask someone to explain everything, accept a label, or promise a long-term outcome. It helps make room for one safer next step.

One moment, not two silos

Substance-use pressure and emotional distress frequently arrive together, but product categories often separate them. One app tracks sobriety. Another tracks mood. A crisis page waits behind a menu. A generic chatbot asks the user to narrate a situation they may not have the energy or privacy to describe.

MARGIN begins with the felt moment instead:

• A craving
• Too much at once
• Feeling low
• Feeling anxious
• Feeling alone
• Not sure

“Not sure” is treated as a complete answer. The product then asks what “safer” should mean for the next few minutes: create a pause, change the immediate environment, bring another person closer, or open urgent support.

The result is a three-part margin chosen by the user:

1. One action for the body
2. One change to the surrounding space
3. One human connection

The actions are deliberately small. MARGIN does not promise that a craving will disappear or that a grounding step will resolve a mental-health condition. Its job is to reduce the size of the next decision.

The quiet relay

Reaching out can be harder than knowing that support might help. A person may not know what to say, may fear becoming a burden, or may not want to explain the entire situation.

The quiet relay provides a short, editable message:

I’m having a high-pressure moment. Can you stay with me for ten minutes? You don’t need to fix anything.

It is an invitation, not an automatic alert. MARGIN does not choose a contact, send a message, monitor the response, or imply that someone is watching the session.

Designing against shame

The prototype contains no sobriety clock, recovery score, red failure state, or lost streak. A return to use cannot erase earlier work because MARGIN does not turn recovery into a game board.

The reflection at the end asks only what changed: a little easier, about the same, or harder. “About the same” is not failure. “Harder” opens a route toward more support. The language centers agency and person-defined positive change rather than compliance.

Safety is part of the interface

Urgent Help is always visible. The dialog separates situations that should not be collapsed into one generic crisis button:

• A possible overdose or medical emergency
• Emotional distress or a substance-use crisis
• Treatment and referral support
• Withdrawal that may require medical care

For U.S. users, the prototype links directly to 911, 988, the SAMHSA National Helpline, and FindTreatment.gov. It also states what the product cannot do: MARGIN does not monitor the page, contact emergency services, diagnose a condition, recommend medication, or provide a self-detox plan.

The experience is a concept prototype, not a clinically evaluated intervention or a substitute for emergency, medical, or professional care.

Privacy by subtraction

No account is required. Selections remain only in the current page memory and disappear when the page is closed or refreshed. The prototype uses no local storage, session storage, advertising identifiers, analytics, pixels, or session replay.

The privacy explanation does not call the experience anonymous. Ordinary hosting infrastructure may retain request or security logs. A quick-exit action replaces the current history entry when it opens a neutral weather page, while clearly warning that it cannot erase other browser, device, network, or download traces.

A directed agent team

I bring lived experience to MARGIN as a person in recovery from addiction and alcoholism, together with a trauma-informed perspective. That experience shapes the product’s emphasis on agency, non-stigmatizing language, privacy, and one manageable next step. It is one person’s perspective, not a substitute for broader co-design or professional review.

I built MARGIN with a team of specialist AI agents assigned to distinct responsibilities: clinical-safety research, addiction-informed product strategy, trauma-informed UX and accessibility, and visual identity. Their work was reviewed and integrated into one product system rather than presented as an automated consensus.

That process is not a substitute for human governance. Before any production or public-service pilot, MARGIN would require paid co-design authority for people with lived and living experience, plus formal review by addiction and mental-health clinicians, crisis specialists, privacy and legal counsel, and disability-accessibility participants.

What the prototype proves

MARGIN does not prove a health outcome. It demonstrates a product thesis:

A support tool can be useful without demanding an identity, collecting a history, maximizing engagement, or scoring recovery.

The prototype is intentionally easy to leave. Success is not time on site. Success is whether the interface helps someone find one manageable next move—and gets out of the way.