5
Programs, each reported separately.
Programs are reported separately and are never ranked, scored against each other or ordered by a result. Each program is measured against its own evidence.
Where a competency holds across a program and where it does not. Every cell is a count of interactions the engine already graded at the moment they happened. Nothing here names a member of staff and nothing here quotes one.
Stored demonstration workforce. Read from the database. Every session below is a stored row, and every reading on this page is computed from those rows by the same engine that graded them. The staff are fictional and the organization is marked as a demonstration, which is what makes this readable without signing in.
Counted here and in no real report. Every record counted here was fabricated to demonstrate the platform. None of it is evidence about anybody, and none of it is counted in any report about a real program.
What this view will not do
Programs, each reported separately.
Programs are reported separately and are never ranked, scored against each other or ordered by a result. Each program is measured against its own evidence.
Trained staff before any share is shown.
Below that a share is a roster. In a small unit a leader who reads that four of six are below standard knows which four. The arithmetic is anonymous and the room is not.
Sessions behind this page, read from the database.
52 members of staff across 5 programs. Every reading is counted from the graded turns underneath them.
Programs held back by that floor today.
Held back rather than estimated, and named below so the gap is visible. A program dropped from this page would read as a program with nothing to report.
PRTF
Aggregated from 15 members of staff in PRTF. No individual record is shown in this view, and a competency with fewer than 10 measured staff is left unscored rather than estimated. Every record counted here was fabricated to demonstrate the platform. None of it is evidence about anybody, and none of it is counted in any report about a real program.
At standard is the share of measured staff whose reading in that competency is at standard. Live gap counts staff carrying an unresolved pattern in it. Transferred counts staff who have demonstrated it across several young people and age groups.
Where this program’s training money goes first: Boundaries, Limits & Consistency. 40% of 15 measured staff read at standard on it.
Targeted practice measurably improved the thing it targeted in 1 of 1 measured cases.
Day Treatment
Aggregated from 11 members of staff in Day Treatment. No individual record is shown in this view, and a competency with fewer than 10 measured staff is left unscored rather than estimated. Every record counted here was fabricated to demonstrate the platform. None of it is evidence about anybody, and none of it is counted in any report about a real program.
At standard is the share of measured staff whose reading in that competency is at standard. Live gap counts staff carrying an unresolved pattern in it. Transferred counts staff who have demonstrated it across several young people and age groups.
Where this program’s training money goes first: Boundaries, Limits & Consistency. 64% of 11 measured staff read at standard on it.
Therapeutic Foster Care
Aggregated from 10 members of staff in Therapeutic Foster Care. No individual record is shown in this view, and a competency with fewer than 10 measured staff is left unscored rather than estimated. Every record counted here was fabricated to demonstrate the platform. None of it is evidence about anybody, and none of it is counted in any report about a real program.
At standard is the share of measured staff whose reading in that competency is at standard. Live gap counts staff carrying an unresolved pattern in it. Transferred counts staff who have demonstrated it across several young people and age groups.
Where this program’s training money goes first: Boundaries, Limits & Consistency. 50% of 10 measured staff read at standard on it.
The Relatives
Aggregated from 10 members of staff in The Relatives. No individual record is shown in this view, and a competency with fewer than 10 measured staff is left unscored rather than estimated. Every record counted here was fabricated to demonstrate the platform. None of it is evidence about anybody, and none of it is counted in any report about a real program.
At standard is the share of measured staff whose reading in that competency is at standard. Live gap counts staff carrying an unresolved pattern in it. Transferred counts staff who have demonstrated it across several young people and age groups.
Where this program’s training money goes first: Active Listening & Emotional Validation. 50% of 10 measured staff read at standard on it.
Held back
These programs are running and their staff hold individual records. What cannot be shown is a share across the team, because a share across a team this small identifies people.
not enough people to report
6 of the 10 needed. The floor is one number in one file and an organization may raise it. Nothing may lower it.
Who may see this
Their organization in aggregate, plus configuration, and the audit trail of who viewed or changed its records.
The staff assigned to them, at their own site.
Named programs, in aggregate only. Never a person, a transcript or a score.
Their own record. Nobody else's, in any view.
An advisor is a consultant, a board member or a funder’s reviewer. Somebody with a real interest in how a program is doing and no standing in how one employee is doing. The refusal is three separate functions in src/flight/access.ts, and each one answers on the role before it looks at any list. An advisor account handed a list of staff ids by mistake still opens nothing.
What is true of this page today
Every share, refusal and cell above was computed by the same code that would run on real training. The staff underneath are fictional.
These are stored rows. The page reads them through the same row-level security as every other reader, because the organization carries a demonstration flag rather than because the read has an exception in it.
The schema, the policies and the seed live in supabase/. A real viewer’s page is scoped by those policies. An advisor reaches the shares and never a person, and that refusal is a query returning nothing rather than a page choosing not to render.
One person’s record is here, and the standards matrix is here.