Read-only inspection
`cuda-doctor check`
check is the lower-cost probe for automation and fast triage when you do not need a full repair workflow.
Where check fits
- CI pipelines that need a machine-readable readiness signal.
- Pre-flight checks before expensive validation or build work.
- Read-only scripting where mutation is not allowed.
- Quick triage on shared systems where a full doctor pass is unnecessary.
Output model
The output should stay concise but still flag hard blockers such as missing architecture support, obvious driver/runtime mismatches, and build-chain gaps that make the machine non-viable for the target GPU.
CI-friendly
Machine-readable output matters here. check should be easy to parse for status dashboards, gatekeeping, and automated rollouts.
Related docs
Diagnose
doctor
Run a full environment diagnosis for the GPU, driver, toolkit, runtime, build chain, and validation risk.
Execution
validate
Prove that device selection, memory transfer, kernel launch, and runtime behavior work on the local GPU.
Execution
build
Build CUDA code in the current project with the correct toolkit, compiler, and architecture settings for the local machine.