Evaluation coverage

CloudEval evaluates infrastructure against 1,700+ cloud evaluation signals across supported Azure and AWS workflows. That platform signal floor includes CloudEval project modeling, architecture graph context, cost and FinOps context, reviewed scoring, deployment quality, IaC findings, report-readiness checks, release gates, and attributed third-party evidence.

Applicable checks depend on provider, IaC format, resources, project configuration, and enabled policies. Not every project runs every check.

Summary

Metric	Current public floor
Cloud evaluation signals across supported workflows	1,700+
Attributed third-party catalog across pinned versions	1,650+ scanner checks
Azure ARM workflow third-party catalog checks	743
Azure Bicep workflow third-party catalog checks	684
AWS CloudFormation workflow third-party catalog checks	460
AWS Terraform workflow third-party catalog checks	455
Azure Terraform workflow third-party catalog checks	302
Cost intelligence coverage	Common billable Azure services
Architecture graph intelligence	Included

Workflow counts overlap and must not be added together. Use the 1,700+ signal count as the stable CloudEval platform coverage floor.

How signals become report output

What the counts mean

Count	Meaning
`1,700+ cloud evaluation signals`	The published lower-bound count for CloudEval platform coverage across supported architecture, security, deployment-quality, cost, graph, report, gate, and IaC evaluation paths.
`1,650+ scanner checks`	The published lower-bound count for attributed third-party tool catalogs across pinned versions.
Workflow counts	Views into attributed third-party catalogs for a provider and IaC format. These overlap.
Findings	Project-specific observations that appear only when the project has matching resources, files, configuration, and enabled policies.
`Not assessed`	A clear status for unsupported or out-of-scope provider, format, resource, or policy areas.

Category breakdown

Category	Current treatment
Azure architecture review	Reviewed Azure architecture mappings from supported Azure workflows.
AWS architecture review	Beta treatment for reviewed AWS CloudFormation controls where mapped.
Deployment quality	ARM TTK and cfn-lint observations, plus CloudEval deployment-readiness checks where applicable.
Additional IaC findings	Checkov observations for supported IaC formats.
Cost intelligence	Estimates and recommendations for common billable Azure services.
Architecture graph intelligence	Graph context used to reason about topology and review context.
Diagram integrity	Checks that keep generated diagrams parseable and reviewable.

Evaluation timing

For a normal-sized template project, the first architecture view is typically ready in seconds. Reports and findings take longer because they evaluate more of the project. Eligible evaluations can run at the same time, so the values below are readiness windows for each stage, not steps in a serial pipeline.

Stage	Typical readiness (p50)	Most ready by (p95)
Architecture diagram	12 seconds	102 seconds
Initial reports	65 seconds	131 seconds
Security findings	75 seconds	162 seconds
Cost analysis	29 seconds	62 seconds
Repository import	78 seconds	152 seconds
Unit tests, when configured	39 seconds	87 seconds

The charts compare each stage from a common starting point. They are visual estimates for a normal-sized template project, not a fixed execution sequence.

Typical readiness (p50)

Half of observed runs reached each stage within the time shown below.

Most ready by (p95)

Ninety-five percent of observed runs reached each stage within the time shown below.

These figures are indicative observations, not an SLA. Timing varies with template size, repository complexity, enabled evaluations, and current demand. Optional stages appear only when they are configured for the project.

Coverage boundaries

The 1,700+ cloud evaluation signals count is a CloudEval platform count, not a scanner-rule total.
Not every project runs every check.
Workflow-specific third-party counts overlap.
AWS CloudFormation support is beta and is not an end-to-end AWS Well-Architected assessment.
Checkov observations appear as additional IaC findings unless CloudEval has reviewed a different output treatment.
Third-party projects are independent; no affiliation, certification, sponsorship, or approval is implied.

Trust and limits

Platform reference

Summary

How signals become report output

What the counts mean

Category breakdown

Evaluation timing

Typical readiness (p50)

Most ready by (p95)

Coverage boundaries

​Summary

​How signals become report output

​What the counts mean

​Category breakdown

​Evaluation timing

​Typical readiness (p50)

​Most ready by (p95)

​Coverage boundaries

​Related pages

Summary

How signals become report output

What the counts mean

Category breakdown

Evaluation timing

Typical readiness (p50)

Most ready by (p95)

Coverage boundaries

Related pages