Deskrune Methodology · Issue 01

How we build a kit.

Every kit on this site starts the same way: with a failure mode, not a feature list. A buyer comes in with one specific moment that keeps breaking — re-entry after a gap, a hyperfocus crash, a project they can't reopen — and the kit is the protocol for that moment. This page is the working document that says how that gets built and what we will and will not put on the cover.

Published9 May 2026 Reading time9 minutes Audit cadenceQuarterly Refund14 days, no questions
01 INPUT 02 MAP 03 MECH 04 KIT failure pattern one of ten AFMs cited mechanism the protocol DESKRUNE / DIAGNOSTIC FLOW

Fig. 1 — every kit travels this path before it gets a cover.

Sections
  1. The taxonomy
  2. The diagnostic
  3. What we cite
  4. What we don't claim
  5. The audit cycle
  6. The refund

01/ 06The taxonomy

Ten ways a system actually breaks.

Most ADHD products start from a subtype label — inattentive, hyperactive, combined — and try to wrap a system around the label. That isn't useful when the issue is that on a given Wednesday, an open Claude project no longer parses as anything an ADHD adult recognizes. The label didn't break. The system did.

So kits are sorted by the failure mode that keeps coming back. Ten of them, the AFMs (AI Failure Modes). Each one is a named moment with observable signals, a working hypothesis about what's going on cognitively, and a small protocol for the moment when the moment happens. The full taxonomy lives at the library; each AFM has its own page with the signals and the circuit breaker.

Sorting by failure mode does two useful things. It tells a buyer in one sentence whether a kit is for them. And it forces every section of the kit to defend its place — if a worksheet doesn't address the named failure mode, it doesn't ship.

Kits aren't sorted by ADHD subtype. They're sorted by the failure mode that keeps coming back.

— The taxonomy rule

02/ 06The diagnostic flow

From your moment to a protocol that fits it.

The flow has four stages. They run the same way for a kit going into the catalog and for a buyer trying to find the one that maps to what's happening in their week.

STAGE 01 STAGE 02 STAGE 03 STAGE 04 INPUT AFM MAP MECHANISM PROTOCOL the failure pattern "this keeps happening" one of the ten AFMs named, with signals cognitive mechanism cited where citable the kit $4.99 PDF, plain, printable DESKRUNE / METHODOLOGY / FIG. 2 "diagnose, then prescribe"

Fig. 2 — diagnostic flow, four stages, no shortcut.

Input. A pattern, not a feeling. "I keep losing the project mid-week" is input. "I need a productivity system" isn't — that's a label, and the diagnostic doesn't run on labels.

AFM map. The pattern gets matched to one of the ten failure modes by signals — observable behaviors, not self-report scales. Re-Entry Collapse has its own signals; Hyperfocus Tunnel has different ones. Two AFMs at once is normal. The kit names which one it's actually built for.

Mechanism. The cognitive mechanism behind the failure mode, cited where the literature is solid (working memory, temporal discounting, prefrontal under-recruitment) and named as a working hypothesis where it isn't. The full citation block is public.

Protocol. The kit. A short PDF, printable, no app, no login, no streak counter. It's the smallest amount of paper that handles the moment.

03/ 06What we cite

The studies that hold a claim up.

A claim about ADHD only goes in a kit if there's a citation that backs it, or it's marked as a pattern observation and labeled as one. Below is the foundational literature kits draw from. The full block — every PubMed ID, every retraction we've watched for — sits at /research/.

  • Stimulant efficacy Cortese et al., 2018. Network meta-analysis of pharmacological treatments for adult ADHD. The Lancet Psychiatry. PMID 30097390. Anchors any claim a kit makes about medication response rates.
  • Sleep / DSPS Bijlenga et al., 2019. Delayed sleep-phase syndrome and ADHD circadian patterns. Anchors the night-shift and chronotype kits.
  • Executive function Barkley, 1997 onward. The temporal self-regulation and behavioral inhibition model. Tier B — clinically respected synthesis built from peer-reviewed work; cited as a model, not a primary trial.
  • Working memory Patros et al., 2016. Working-memory deficits in ADHD; meta-analytic review. Used wherever a kit references the 4-second working-memory window.
  • Lifespan persistence Hechtman et al., 2016 (MTA follow-up). Long-term persistence of executive-function deficits into adulthood. Used by the late-diagnosis and re-entry kits.
  • Dopamine signaling Volkow et al., 2009 onward. Reward-circuit and dopamine-signaling differences in ADHD adults. Used where a kit discusses task-initiation friction.
  • RSD framing Surman et al., 2011. Emotional-dysregulation prevalence in adult ADHD, the most rigorous source for the AFM-6 framing. RSD as a clinical construct is still being formalized; the kit copy says so.
If we wouldn't ship a kit on retracted research, we won't keep selling one that becomes that.

— The retraction rule

04/ 06What we don't claim

What a kit isn't.

Equally important. A buyer should know what's not in the box before they pay $4.99. Cross-checked at /honest/.

  • Not medical advice.

    Kits describe patterns and protocols. They aren't a treatment plan and don't replace one. Decisions about medication, diagnosis, or care belong with a clinician.

  • Not a substitute for a clinician.

    A worksheet for re-entry after nine days offline is useful. It is not a therapist, a psychiatrist, or an ADHD coach. If a buyer needs one of those, the kit says so on the cover.

  • Not a fix for ADHD.

    Nothing on this site claims to fix ADHD. The kits help with specific moments. The brain is the brain.

  • Not a personality test.

    No quizzes that sort buyers into archetypes. The kits map to failure modes, not to a buyer's identity. Sorting people is what the rest of the productivity industry does. The team isn't doing that here.

  • Not gamified.

    No streak counters, no XP, no shame-on-Tuesday-because-you-skipped-Monday badges. Streaks hurt the people the kits are for. They are banned across the catalog on a hard rule.

05/ 06The audit cycle

What gets re-checked, every quarter.

Every kit goes through a validation pass four times a year. Three things get checked: the citations, the AI behavior the kits depend on, and the voice canon.

Citations. Every PubMed ID on every kit page resolves. If a study has been retracted or superseded, the kit gets pulled or revised, and the change shows up in the public revision log within seven days of discovery.

AI behavior. Kits that reference how Claude, ChatGPT, or Gemini respond to a specific prompt structure get re-tested. Models drift. A prompt that worked in March on one model won't necessarily produce the same answer in July, and a kit built around a stale prompt is a kit selling a broken protocol. When the prompt regresses, the kit is updated.

Voice. Every page is run against the banned-word list. The list is short and public. A page that fails goes back to draft.

Buyers get every revised version free. No second purchase, no email gauntlet — once a kit is bought, every quarterly update lands in the same downloads folder. The revision log is at /changes/; nothing ships in private.

06/ 06The refund

Fourteen days, no questions.

A buyer who opens a kit and finds it isn't the right one for them emails [email protected] within fourteen days. The money goes back. There's no form, no exit-interview survey, no "tell us why" gate. The team trusts the buyer to tell the truth, and the price doesn't make a defensive policy worth the friction.

One more thing on price. There are no discount codes, no "limited tier of 100," no exit-intent overlays. The price is $4.99 through May 31, and on June 1 it's $4.99. The end date is on the buy page because the end date is real.

We trust you. Here's the work.

See the kits — $4.99 through 31 May

Deskrune. For when you come back.