Skip to content

sodejm/network-segmentation-workbench

v0.1.0PolyForm-Noncommercial-1.0.0

Compare firewall and workload policy intent with flow, DNS and bounded laboratory reachability evidence.

COPS Network Segmentation Workbench

Compare firewall and workload policy intent with flow, DNS and bounded laboratory reachability evidence.

Maturity: experimental. Offline correction playback is available. Laboratory isolation, host installation and production integrations remain unverified.

Usage

From the repository, inspect the package metadata, run its fixture playback, and validate its package contract:

python3 -m cops info network-segmentation-workbench
python3 -m cops demo network-segmentation-workbench
python3 -m cops check network-segmentation-workbench

From an exported package, run python3 scripts/demo.py. Inspect supplied input with python3 scripts/assess.py --input examples/input.json --as-of 2026-10-15T12:00:00Z.

Create a strict native report with python3 scripts/report.py --input examples/input.json --as-of 2026-10-15T12:00:00Z --report-id report-demo --engagement-id eng-demo --scenario-id COPS-DEMO --result-id res-demo --target fixture:network-demo. The caller must supply those context identifiers; the wrapper does not invent engagement or target provenance.

Design

The package uses the bounded COPS artifact assessment engine and content-free SHA-256 evidence references. An immutable copy of its runtime is included in vendor/runtime; update it using the repository runtime synchronization script. The assessment does not establish production exploitability or compliance.

Data Contract

ArtifactPurposeProducerConsumerValidation
examples/input.jsonSupplied artifact envelopeAuthorized collectorAssessment engineType and byte bounds
examples/correction-cases.jsonCorrection acceptance casesSynthetic fixture authorDemo and package checksExpected codes and content hashes
Assessment resultFindings, coverage and limitationsEngineReviewer or report exporterNative Finding contract and SHA-256
Assessment reportStrict imported-evidence reportscripts/report.pyNative report exportercops.assessment-report/v1 semantic validation

See the runnable playbook, verification, and walkthrough.

Typed policy and observation inputs

Run python3 scripts/assess.py --input examples/typed-input.json --as-of 2026-10-15T12:00:00Z for corrected first-match firewall, additive NetworkPolicy, explicit flow and TTL-bound DNS examples. scripts/demo.py also checks their vulnerable and insufficient counterparts. These synthetic snapshots establish parser behavior; HTTP authorization denials do not establish network enforcement.