Message leg of the truthful top-X/N-hours report: hub-message-census.sh, rows and sends side by side with the aro half declared unavailable

KPI-80 kpi Done high ↳ KPI-66

Fleet token monitor cutover and old-stack retirement

KPI-74 kpi Done high

└─ Accept fleet ranking API and simple web page

KPI-78 kpi Done high

└─ Bring Lezama Claude and Codex token feed live

KPI-77 kpi Done high

└─ Bring Whey Claude and Codex token feed live

KPI-76 kpi Done high

└─ Bind ingest credentials to Venus, Whey and Lezama hosts

KPI-75 kpi Done high

└─ Retire remaining old tracking stack and references

KPI-79 kpi Done

Collection: coverage tracking + automation_job class

KPI-67 kpi Done high ↳ KPI-66

kpi-burn-alert.sh cannot see unregistered codex agents - read v_usage_resolved, not usage_events

KPI-72 kpi Done high

Fleet-wide usage dashboard on kpi.pensanta.com: every agent, every host, both providers, ranked

KPI-58 kpi Done high ↳ KPI-45

└─ KPI-58 retire the WEB/EXPORT read of token-usage.sqlite + verified snapshot (collector stays)

KPI-61 kpi Done high

└─ KPI-58 whey ingestion: usage_events has zero whey rows

KPI-62 kpi Done high

└─ KPI-58 web cutover: token-usage-web.sh /, /export.json, /export.md read Postgres usage_events (pg_rank_sql contract)

KPI-60 kpi Done high

└─ KPI-58 fleet notice: export.json agent set doubles at pg cutover

KPI-63 kpi Done high

fleet-tokens.sh - one-command top-offenders / per-agent token answer (monitor-context-cc)

KPI-59 kpi Done high ↳ KPI-45

INCIDENT (done): 19h fleet-wide compact blind spot from CC statusline format change

KPI-57 kpi Done high ↳ KPI-45

kpi-web (whey :9876, ONEDB-5): 17-view telemetry dashboard - maintainer nw-venus-cc, state contested

KPI-42 kpi Done ↳ KPI-45

Lezama token-usage collector (no cross-host ingest exists)

KPI-50 kpi Done low ↳ KPI-45