Inventory and change tracking
A searchable estate catalogue enriched with cost and utilization, and exact diffs answering what changed between two points in time.
Inventory #
Every resource in scope, normalized into a consistent shape and enriched with what you usually have to join manually:
- Identity, type, subscription, resource group, region, SKU, and tags
- Actual and amortized monthly cost
- Utilization and its telemetry source
- Any open opportunity against the resource
Filter by search term, type, subscription, or region. Counts are explicit — visible versus total — so a filtered view is never mistaken for the whole estate.
Saved views store filter combinations you return to, and filters are carried in the URL, so a filtered view can be shared as a link.
Exports #
Export the filtered set to CSV or Excel. Excel carries a metadata sheet with generation time, snapshot version, and applied filters. CSV is protected against spreadsheet formula injection.
Resource detail #
Opening a resource shows its full context: cost history, per-metric telemetry with source and window, tags, and any findings against it — the whole picture without visiting three other tools.
Changes #
Inventory snapshots are retained, so Flux can answer what changed as an exact diff between two consecutive snapshots rather than a guess.
Detected change types:
| Change | Example |
|---|---|
| Created | A new resource appeared |
| Deleted | A resource is gone |
| Resized | SKU or capacity changed |
| Re-tiered | Storage or service tier changed |
| Re-tagged | Tags added, removed, or edited |
| Moved | Resource group or subscription changed |
| Reconfigured | Other tracked configuration changed |
Each carries the before and after values, so a resize shows what it was and what it became.
[!TIP] The question this answers “Cost jumped on Tuesday” plus “these fourteen things changed on Tuesday” is usually a five-minute investigation instead of a half-day one. Pair this with cost anomalies.
Change anomalies #
Change volume is itself monitored. Flux builds a median/MAD baseline of how much normally changes per snapshot and flags unusual bursts.
A day with two hundred deletions when thirty is normal is worth knowing about regardless of cost impact — it may be a migration, or it may be a mistake.
As elsewhere, a thin baseline reports warming up rather than firing.
Governance #
Policy compliance posture across subscriptions and assignments, with evaluated, compliant, non-compliant, and exempt counts, plus a resource-level drilldown naming the specific definition each resource violates.
Exemptions are shown rather than hidden, since an exemption is a decision someone made and should be reviewable.