slug: runai provider: NVIDIA Run:ai generated_by: planning/capability-mapping/scripts/classify_capabilities.py model: claude-opus-5 frame: - Electrical Components & Equipment min_confidence: 0.7 capability_model: source: https://github.com/vincentmakes/turbo-ea-capabilities license: CC-BY-4.0 attribution: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities, CC BY 4.0 notice: NOTICE edge_count: 19 edges: - tag: Access rules spec_file: runai-access-rules-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.85 evidence: POST /api/v1/authorization/access-rules create_access_rule; schemas AccessRuleSubjects, ScopeType, ScopeId reason: Authorization rules binding subjects to roles at given scopes — plainly identity and access management. recovered_from: sweep-20260829T005356Z-edges.json - tag: Idps spec_file: runai-idps-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.85 evidence: POST /api/v1/idps 'Configure external idp'; 'Get idp mappers'; schemas 'SamlCreationData', 'OidcData', 'MandatoryClaim' reason: Configuration of external SAML/OIDC identity providers and claim mappers for platform login — identity federation, squarely Identity & Access Management. - tag: Roles spec_file: runai-roles-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.85 evidence: GET /api/v1/authorization/roles 'Get a list of roles.'; schemas Permissions, KubernetesPermissions, RoleV1 reason: Authorization roles and permission sets used to grant access to platform resources — squarely identity and access management. - tag: Distributed spec_file: runai-distributed-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.8 evidence: POST /api/v1/workloads/distributed 'Create a distributed training.'; 'Suspend a distributed training.'; schema 'DistributedSpec' reason: Operations create, suspend, resume and delete distributed training workloads on GPU clusters — this is ML training lifecycle / MLOps execution, mapping to Artificial Intelligence Management. - tag: Distributed Inferences spec_file: runai-distributed-inferences-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.8 evidence: POST /api/v1/workloads/distributed-inferences 'Create a distributed inference. [Experimental]'; 'Update distributed inference spec.' reason: Lifecycle of distributed model-inference workloads (serving) on GPU infrastructure — model deployment/serving is part of the AI/ML model lifecycle and MLOps capability. - tag: Inferences spec_file: runai-inferences-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.8 evidence: POST /api/v1/workloads/inferences 'Create an inference.'; 'Get inference metrics data.'; schema 'InferenceSpecSpec' reason: Deployment, update and metric monitoring of model-inference workloads on GPU clusters — model serving within the AI/ML model lifecycle and MLOps capability. - tag: Permissions spec_file: runai-permissions-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.8 evidence: GET /api/v1/authorization/permissions "Get a summary of user permissions."; POST /api/v1/authorization/permitted-scopes "Calculate permitted scopes" reason: Operations resolve user permissions and permitted scopes under an /authorization path with Permission/Action/ResourceType schemas — authorisation evaluation, which is Identity & Access Management. recovered_from: sweep-20260829T005356Z-edges.json - tag: Service Accounts spec_file: runai-service-accounts-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.8 evidence: POST /api/v1/service-accounts 'Create a service account.'; POST '/service-accounts/{serviceAccountId}/secret' 'Regenerate a service account secret.' reason: Lifecycle of non-human identities and their credentials used to authenticate to the API — machine identity and credential management within IAM. - tag: Access Keys spec_file: runai-access-keys-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.78 evidence: POST /api/v1/access-keys create_access_key1 Create an access key; POST /api/v1/access-keys/{accessKeyId}/secret regenerate_access_key_secret reason: Issuance, rotation and administrative deletion of user/service-account access keys used to authenticate to the control plane — identity and access credential management. recovered_from: sweep-20260829T005356Z-edges.json - tag: Clusters spec_file: runai-clusters-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.75 evidence: POST /api/v1/clusters create_cluster Create a cluster.; GET /api/v1/clusters/{clusterUuid}/metrics get_cluster_metrics Get the cluster metrics data.; GET /v1/k8s/clusters/{cluster_uuid}/installfile getInstallFile reason: Operations create, update, install and monitor Kubernetes GPU clusters — management of compute infrastructure, i.e. IT Infrastructure Management. Some overlap with IT Operations monitoring keeps confidence below 0.8. - tag: Trainings spec_file: runai-trainings-api-openapi.yml capability_id: BC-610.60 capability_id_l1: BC-610 capability_name: Artificial Intelligence Management confidence: 0.75 evidence: POST /api/v1/workloads/trainings 'Create a training.'; 'Suspend a training.'; 'Resume a training.' reason: Lifecycle control of model training jobs on a GPU orchestration platform is ML workload/MLOps management, the closest candidate being Artificial Intelligence Management. - tag: Users spec_file: runai-users-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.75 evidence: POST /api/v1/users 'Create a local user.'; POST /api/v1/users/{userId}/password 'Reset a user's password.' reason: User account provisioning, password reset and deletion within the platform is Identity & Access Management, not HR employee records. - tag: Applications spec_file: runai-applications-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.72 evidence: POST /api/v1/apps create_application; POST /api/v1/apps/{appId}/secret regenerate_application_secret Regenerate an application secret. reason: '''Applications'' here are machine identities with client secrets used for API authentication (the description ties them to service accounts and access rules), i.e. credential/identity management rather than application portfolio management.' recovered_from: sweep-20260829T005356Z-edges.json - tag: Storage Class Configuration spec_file: runai-storage-class-configuration-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.72 evidence: POST /api/v1/storage-class-configuration 'Create storage class configuration'; schemas PvcAccessModes, PvcVolumeMode, StorageClassPermissions reason: Configuration of Kubernetes storage classes and PVC attributes for the GPU cluster is storage/cloud infrastructure administration, i.e. IT Infrastructure Management. No business-domain reading fits. - tag: Storage Classes spec_file: runai-storage-classes-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.72 evidence: GET /v1/k8s/clusters/{uuid}/storage-classes 'Get all storageClasses from a cluster.' reason: Read surface over cluster storage classes — infrastructure (storage) inventory for a Kubernetes platform, mapping to IT Infrastructure Management. - tag: Compute spec_file: runai-compute-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.7 evidence: POST /api/v1/asset/compute create_compute_asset Create compute asset.; schemas GpuDevicesRequest, CpuCoreRequest, CpuMemoryLimit reason: Defines reusable compute resource specifications (CPU cores, memory, GPU devices) allocated to workloads — compute infrastructure definition. Confidence moderate because these are reusable asset templates rather than physical infrastructure operation. - tag: Credentials spec_file: runai-credentials-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.7 evidence: Create an access key. / Create a userid / password credential. / List docker registry credentials. reason: Operations manage access keys, passwords, docker-registry credentials and generic secrets — credential/secret stewardship, i.e. identity & access management. Could also be read as configuration & secrets management, hence moderate confidence. recovered_from: sweep-20260829T005356Z-edges.json - tag: Me spec_file: runai-me-api-openapi.yml capability_id: BC-620.20 capability_id_l1: BC-620 capability_name: Identity & Access Management confidence: 0.7 evidence: GET /api/v1/authorization/me/access-rules 'List the access rules assigned to the requesting user.' reason: Introspection of the calling user's authorization rules and scopes — access-rights resolution within Identity & Access Management. Read-only self-service endpoint, so confidence moderate. - tag: NodePools spec_file: runai-node-pools-api-openapi.yml reanchored_from: runai-nodepools-api-openapi.yml capability_id: BC-600.50 capability_id_l1: BC-600 capability_name: IT Infrastructure Management confidence: 0.7 evidence: POST /v1/k8s/clusters/{clusterId}/node-pools createNodePool Create a Node Pool. reason: Lifecycle and metrics of Kubernetes node pools (compute capacity groupings) within clusters — compute/cloud infrastructure management, not any industry business capability.