For policymakers, counsel, agencies and funders

Five frameworks, one argument.

Each framework is paired with working software that makes it operable rather than theoretical. None of them asks for a new appropriation, a new agency, or a rule change before it can start working. All of them start from the same finding: the programs largely exist, they are funded, and the people who need them cannot see them from where they are standing.

This page is the index to the written work — the models, the funding stacks, the model bills, the metrics, and the conditions that would prove each one wrong.

Looking for the applications themselves?

The people these are built for do not come through this page. Each audience has its own entrance, in its own language, with nothing on it that belongs to somebody else.

Veterans → SSDI and SSI → Correctional and recovery →

The series

Written in order. Each one inherits the design rules of the one before it.

	Framework	What it covers
01	CornerstoneCorrections	Correctional campuses. Turns the time between sentencing and release into structured training that produces a credential, a job offer, treated clinical need, and a verifiable record. Five applications.
02	DrydockRecovery	Treatment and recovery as infrastructure rather than as a gate. Four phases from court through work release. Supplies the design rule the other three inherit: recovery is the path with the rewards on it, never the condition in front of the benefit.
03	StandfastVeterans	Purpose, healing, and safe places in the year after separation, with no discharge-status gate — because the highest-risk group is disproportionately the one a status gate would turn away. Work is treated as the win where a veteran wants it, and never as the measure of whether they count.
04	LintelDisability	Makes the benefits cliff legible, leads with what is protected, and guarantees the way back. SSDI has a genuine cliff; SSI has none at all, and almost nobody in either program is told the difference.
05	HoldfastCaregivers	Veteran caregivers and the survivor transition. The program gives 150 days of runway when the veteran improves and 90 when the veteran dies — the only exit in which the person is bereaved, unemployed, and losing their health coverage at once. Ranks every route out by the runway it actually leaves.

The rules every one of them follows

These are load-bearing. Any adaptation that drops one of them is not this model.

Recovery is never a conditionTreatment sits on the path with the rewards on it. It is never the gate a person has to pass before the benefit is available to them.
Separate doors where the arithmetic differsA veteran routed through a general benefits door gets a wrong number, because VA compensation does not touch SSDI and reduces SSI almost dollar for dollar. Where the rules diverge, the entrance diverges.
Gaps are shown, not filledEvery factual cell is verified, likely, unverified, or open. Only verified items count toward a finding. Lintel's state table displays 4 of 52 rather than 52 plausible numbers, because a wrong threshold would tell someone their health coverage is safe when it is not.
No generated contact detailsThese applications name agencies and programs without phone numbers or URLs. The one deliberate exception across the entire suite is the Veterans Crisis Line, because a wrong number handed to someone in a crisis is worse than no number at all.
Nothing leaves the deviceEvery application is a single file that runs in the browser and transmits nothing. No login, no server, no install — a veteran on a phone should not need an account, and a benefits counselor's office should not need IT approval to open a web page.
Each one states how it could be wrongEvery framework names, in advance and on purpose, the condition that would falsify it.

Watch

The argument in two minutes, and a narrated walkthrough of each application — recorded off the real software, not a mockup, with captions.