Pages
Teams & Users — Strip Levers

Source: qdash /teams — c67 cognitive pass, 2026-08-15

The form is settled; this round questions the information. The current lenses are mostly states — how much, how many. Optimization, improvement, and risk reduction need levers: numbers that move when someone does something on purpose. The test a lens must pass: (a) an unambiguous good direction · (b) attribution to something the team or its admin actually does · (c) a feedback loop shorter than a quarter. Every provenance claim below was verified against qdash source this session.

Audit

The current five, judged

Spendkeeps its slot

A state — and that's fine. It's the accountability ledger, the Cost Center reconciliation point, and the receipts system's spine. One state lens is correct; five was the problem.

$ / persondemoted → tooltip

A normalization of Spend, not a cognitive job — and it fails the direction test: high could be power users or waste. The stat sheet and whisper line already carry it. Its actionable half becomes Waste.

Active nowdemoted → Auto + focus

Weather. Nobody improves active-now; it's refresh-to-refresh noise. But it's the only near-real-time signal — exactly what the Auto lens was built to surface when notable, and what the focus card shows in context.

Agents / persondemoted → focus card

The micro-histogram bake-off finding was real — but the metric has no good direction (depth or sprawl?). A diagnostic, not a directive. The histogram keeps its win inside the focus card.

Risk (open Crit/Elev)upgraded → Open, aged

The surprise of the audit: danger_level is already conditions-based (yolo/sandbox/approval configs), and findings subsume those conditions with a workflow state machine and timestamps. What Risk lacked was a lever — the age of the oldest open item adds one: an SLA someone can hold.

Ledger

Every candidate, with receipts

Provenance: buildable today client-side · derivable from existing records/routes · needs a backend aggregate · the data doesn't exist. The so-what column is the lever test: if this number moves, who did what?

MetricJobIf it moves, who did what?Prov.Verdict
Tokens (native unit)accountabilitythe same ledger in ground-truth units — token counts are measured, dollars are estimatedshipssame by_team rows as Spend; the $/Tokens depth toggle stays on Cost Center (c66)
Waste (idle >14d spend)optimizationa manager re-engaged the idle person or reclaimed the seat/key — either outcome is the job doneshipscosts by_user /30d × rollup last-active; threshold anchored to data-max
Retention (7d cohort return)improvementenablement worked — the people who tried it last month are still hereshipsfirst_seen/last_active from /api/agents (NOT /api/sessions — 200-row cap)
Coverage (instrumented machines)leading risksomeone instrumented a machine — the shadow surface shrank by name, not by vibeshipsinstallations.instrumented via machineRollups; denominator = "machines we can see"
Open, agedlagging risk + SLAsomeone resolved the old finding instead of accumulating new ones — the SLA heldshipsfinding first_seen + workflow statuses; terminal = resolved/accepted
Mapping coveragethe page's own leveran admin sat in the mapping workspace and assigned identitiesverdict linefleet-level, not per-team — it headlines, it doesn't lens
Cache leverageoptimizationprompt/session habits improved — but per-team it mostly reflects harness mix, so movement ≠ the work working→ Cost Centerreachable via /api/costs?team= (the unlock below); cost-composition belongs on /cost
Premium-model shareoptimizationmodel-pick policy changed — but workload mix confounds it→ Cost Centersame ?team= route, by_model rows; per-person split genuinely impossible client-side
Exposure (composed conditions)leading riskkilled2 of 3 components don't exist (no MDM signal in web.rs; no sensitivity classification on top_context); the real third — consumer keys — is /identity's per decision 3. Re-scoring here recreates the dual-formula bug this cycle exists to kill.
Egress surfacerisk reductionallowlist work shrank the destination set — an admin lever, not a team behavior→ controls / focus cardegress_dests is a per-agent count; summing double-counts shared hosts — honest per-team distinct needs a backend aggregate
Guardrail posturerisk reductionpolicy authoring covered more actions — real lever, wrong altitude for a chip→ focus card / controlstool_calls decision_source (config|user|policy) — % policy-covered derivable
Live

The lever strip

One lens per cognitive job: accountability · optimization · improvement · leading risk · lagging risk with an SLA. Each wears its bites-system micro-form; every mark is still a portal. The verdict line carries the page's own lever — mapping coverage — because the strip shouldn't get credit for the workspace's work.

153 of 214 identities assigned · $4.1k of $18.4k still unowned

Teams

ML-Infra$4.2kt.brandt — $69/30d · eleva.lindqvist — $107/30d · activey.sato — $161/30d · activeml-04 — $230/30d · activeml-05 — $314/30d · activeml-06 — $38/30d · activeml-07 — $46/30d · activeml-08 — $69/30d · activeml-09 — $107/30d · activeml-10 — $161/30d · activeml-11 — $69/30d · activeml-12 — $107/30d · activeml-13 — $161/30d · activeml-14 — $230/30d · idleml-15 — $314/30d · idleml-16 — $38/30d · idleml-17 — $46/30d · idleml-18 — $69/30d · idleml-19 — $107/30d · idleml-20 — $161/30d · idleml-21 — $230/30d · idleml-22 — $314/30d · idleml-23 — $38/30d · idleml-24 — $46/30d · idleml-25 — $69/30d · idleml-26 — $107/30d · idleml-27 — $161/30d · idleml-28 — $230/30d · idleml-29 — $314/30d · idleml-30 — $38/30d · idleml-31 — $46/30d · idle23% of fleet · 31 people
Payments-EU$2.9kr.vance — $62/30d · critj.chen — $96/30d · critm.silva — $144/30d · activepa-04 — $206/30d · activepa-05 — $282/30d · activepa-06 — $34/30d · activepa-07 — $41/30d · activepa-08 — $62/30d · activepa-09 — $96/30d · activepa-10 — $144/30d · idlepa-11 — $96/30d · idlepa-12 — $144/30d · idlepa-13 — $206/30d · idlepa-14 — $282/30d · idlepa-15 — $34/30d · idlepa-16 — $41/30d · idlepa-17 — $62/30d · idlepa-18 — $96/30d · idlepa-19 — $144/30d · idlepa-20 — $206/30d · idlepa-21 — $282/30d · idlepa-22 — $34/30d · idlepa-23 — $41/30d · idlepa-24 — $62/30d · idle16% of fleet · 24 people
Core-Banking$600m.varga — $52/30d · critd.osei — $78/30d · critp.novak — $112/30d · elevco-04 — $153/30d · activeco-05 — $19/30d · activeco-06 — $22/30d · idleco-07 — $34/30d · idleco-08 — $52/30d · idleco-09 — $78/30d · idle3% of fleet · 9 people
Research$900a.kim — $114/30d · actives.roy — $170/30d · activere-03 — $243/30d · activere-04 — $332/30d · idlere-05 — $41/30d · idle5% of fleet · 5 people
Support-Ops$300l.novra — $15/30d · activeb.diaz — $23/30d · activesu-03 — $34/30d · activesu-04 — $49/30d · activesu-05 — $66/30d · activesu-06 — $8/30d · activesu-07 — $10/30d · activesu-08 — $15/30d · activesu-09 — $23/30d · activesu-10 — $34/30d · activesu-11 — $10/30d · activesu-12 — $15/30d · idle2% of fleet · 12 people
Unassigned$4.1kj.okafor — $75/30d · critci-pool — $107/30d · activeun-03 — $147/30d · activeun-04 — $18/30d · idleun-05 — $21/30d · idleun-06 — $32/30d · idleun-07 — $50/30d · idleun-08 — $75/30d · idleun-09 — $107/30d · idleun-10 — $147/30d · idleun-11 — $147/30d · idleun-12 — $18/30d · idleun-13 — $21/30d · idleun-14 — $32/30d · idleun-15 — $50/30d · idleun-16 — $75/30d · idleun-17 — $107/30d · idleun-18 — $147/30d · idleun-19 — $18/30d · idleun-20 — $21/30d · idleun-21 — $32/30d · idleun-22 — $50/30d · idleun-23 — $75/30d · idleun-24 — $107/30d · idleun-25 — $147/30d · idleun-26 — $18/30d · idleun-27 — $21/30d · idleun-28 — $32/30d · idleun-29 — $50/30d · idleun-30 — $75/30d · idleun-31 — $107/30d · idleun-32 — $147/30d · idleun-33 — $18/30d · idleun-34 — $21/30d · idleun-35 — $32/30d · idleun-36 — $50/30d · idleun-37 — $75/30d · idleun-38 — $107/30d · idleun-39 — $147/30d · idleun-40 — $18/30d · idleun-41 — $21/30d · idleun-42 — $32/30d · idleun-43 — $50/30d · idleun-44 — $75/30d · idleun-45 — $107/30d · idleun-46 — $147/30d · idleun-47 — $18/30d · idleun-48 — $21/30d · idleun-49 — $32/30d · idleun-50 — $50/30d · idleun-51 — $75/30d · idleun-52 — $107/30d · idleun-53 — $147/30d · idleun-54 — $18/30d · idleun-55 — $21/30d · idleun-56 — $32/30d · idleun-57 — $50/30d · idleun-58 — $75/30d · idleun-59 — $107/30d · idleun-60 — $147/30d · idleun-61 — $18/30d · idle22% of fleet · 61 unmapped
30d attributed spend · same /api/costs rollup Cost Center reads · the amber cloud is the verdict line's unowned $4.1k
Provenance

Receipts, and one correction to our own record

The ?team= unlock — a correction to our own record

/api/costs applies team scoping BEFORE its accumulation loop, so ?team=X&period=30d returns team-scoped totals (including cache %), by_model, and a per-team daily series. Our earlier claim "per-team trends impossible — the daily series is fleet-wide" is true only for a single unscoped fetch. N scoped fetches (N = team count, small pre-SCIM) buys history, cache split, and model split with zero backend work. CYCLE.md corrected; Cost Center's team table inherits the opportunity.

Thresholds are printed, not buried

Waste's 14 days, Retention's 7-day cohort, and Open's age all take a threshold — exactly the kind of judgment the receipts law says to surface. All three anchor to the dataset's MAX(last_seen) (the backend's window_cutoff convention), never wall clock, and the whisper line states the active threshold under every lens.

The joins keep their seams

Waste joins costs by_user (label = email-or-OS-user, lowercased) against the rollup's user key — the same unassigned seam the strip already prints. Coverage's denominator is "machines we can see": an install or an observed agent. No MDM join exists, so the honest gap may be larger than shown — the caption says so instead of smoothing it.

Principles

States vs levers

One accountability job is enough. Spend stays because a page needs its ledger, and Tokens rides beside it as the same ledger in the native unit — token counts are measured where dollars are estimated (the c66 principle). Two units, one job. But four more states were four ways of restating "big team." Every other slot now answers "what should someone do, and how will we know it worked": re-engage or reclaim (Waste), make it stick (Retention), instrument the machine (Coverage), resolve the old one (Open, aged). And the page's own lever — assignment — headlines the verdict line rather than hiding in a lens.

The demotions aren't deletions. $/person lives in the stat sheet, Active-now belongs to the Auto lens (surfaced exactly when notable), the agents-per-person histogram keeps its bake-off win inside the focus card. Metrics moved to where their cognitive job actually is — which is the whole lesson of the pass: the question was never "what can we show" but "what will someone change".

Qpoint Brand Style Guide