Systems Sensor
Watch every system
as it changes
Systems Sensor is the acquisition layer of the ledger — it watches the resources, services, and integrations every initiative depends on, across cloud, identity, endpoint, ITSM, networking, and observability. Current state against declared target, field by field, drift the moment it opens, for every record, every minute.
API-native, no agents
Every connector uses the provider's own API surface. Nothing to install in your environment, nothing to maintain.
Intent-scoped discovery
Panaptico pulls what your declared intent needs. The model stays lean by design — scoped to the operations that matter, not everything with an IP.
One live model
Every discovered object lands in the model as an entity, a signal, a population member, or evidence — ready to be verified.
Watch
148k resources watched, unified across every source
Sensor deduplicates findings from overlapping tools and preserves full provenance for every attribute — so you always know which source said what, when it last refreshed, and whether the record matches target.
148,291 resources unified and continuously reconciled
Deduplicated across sources · current state checked against target
| System Name | System Owner | OS / Distribution | Criticality | Data Source |
|---|---|---|---|---|
| prod-api-gateway-01 | platform@panaptico.com | Ubuntu 22.04 | High |  |
| auth-service-prod-main | IT team | Ubuntu 22.04 | High |  |
| analytics-pipeline-prod | data-eng@panaptico.com | Windows Server 2016 | High |  |
| file-storage-cluster-02 | Infra eng | Centos 7.3.1611 | Medium |  |
| dev-sandbox-testing | curtis.w@panaptico.com | Rocky 8.10 | Medium | |
| nettapphxy-12 | james.p@panaptico.com | Rocky 8.10 | Medium | |
| nettapphxy-13 | james.p@panaptico.com | Rocky 8.10 | Medium | |
Cross-tool deduplication
When Qualys, Orca, and Rapid7 all describe the same host, Panaptico consolidates them into one system record without losing which source observed what.
Full provenance
Every field — owner, criticality, environment, controls — carries the exact data sources that contributed. Drill down to see the underlying reasoning.
Unified and raw views
See the consolidated graph or switch to per-tool raw data. Both are available; the unified view is the default.
Trust
Every attribute, every source, every confidence score
No attribute without provenance. Every value carries which sources said it, how confident the model is, and when it was last checked — so every claim traces back to evidence, not to a dashboard’s word.
tp1dist-01
Ubuntu Linux 22.04
System Metrics
Last updated: 02/20/2026, 12:20:31 AM
| Metric | Value | Data Source(s) |
|---|---|---|
| System Owner | Eng Infra Team | |
| Business Criticality | Critical99% confidence | |
| Internet Facing? | Yes95% confidence | |
| Contains Sensitive Data? | Yes95% confidence | |
| Environment | Production99% confidence | |
| Compensating Controls | Yes98% confidence | |
Cross-tool deduplication
When multiple sources describe the same attribute, Panaptico consolidates them into a single record while preserving the full provenance of which tools observed it.
Confidence scoring on every attribute
See how certain Panaptico's agents are about every judgment. Drill into confidence breakdowns to understand which sources contributed and the reasoning.
Raw and aggregated views
Both unified and source-specific details are preserved. Switch between the single pane of glass and the raw data from each tool at will.
Normalize
Provider-specific objects collapse to shared primitives
The ontology keeps only what verification needs — identity, state, relationships, evidence. Everything else stays in the raw source layer, in your lake, queryable whenever you want it.
Access Principal
Compute Resource
Change Request
Evidence Artifact
Reconcile
Sensor never stops watching
Every category reconciles on its own cadence. When current state diverges from target, Sensor classifies the change — needs review, action required, in progress, or in sync. The status pill earns attention; the noise stays quiet.
Classified, not just alerted
Every drift is categorized — ignorable noise, watched, task-worthy, blocking, approval-gated, or auto-fixable. The engine decides what earns attention.
Continuous refresh
Providers are polled on their own cadence and rate limits. New resources appear in the graph as they show up in reality, not on a scheduled export.
Baselined forever
Once an initiative lands, the model becomes the baseline for drift, readiness, and the next change — without a second onboarding.
Next
Once Sensor sees the truth,
every claim becomes checkable
Gaps between current and declared state become routed work with owners — probes test the fix, signals re-score, and Sensor confirms the claim holds. Met, and staying met.
See the full platform