> ## Documentation Index > Fetch the complete documentation index at: https://docs.getmillwork.dev/llms.txt > Use this file to discover all available pages before exploring further. # Apply tenant template > Create or resume a tenant-template application. ## OpenAPI ````yaml /openapi/solverapi.openapi.json post /v1/tenant-template-applications openapi: 3.1.0 info: description: >- Run approved AI models under data, cost, and time limits. A cost limit can stop another provider call after recorded spend reaches the limit; it cannot reverse a call already in progress. Each finished run returns a receipt naming the selected model and provider without including the prompt or model output. Use the model catalog for what your organization can run now. Provider secrets never appear in API requests or responses. Retry a mutation with the same Idempotency-Key and the same request body. title: Millwork API version: 1.0.0 servers: - url: https://api.getmillwork.dev security: [] tags: - description: >- Submit work, check its status, cancel it, and read the result. The API calls each run an execution. name: Runs - description: A record of how a run was handled, without prompt or result content. name: Receipts - description: Models this organization can use now. name: Models - description: Providers and sign-in methods that can be connected. name: Available providers - description: Provider accounts connected to this organization. name: Provider connections - description: >- One-time browser steps for connecting a provider without sending its secret in an API request. name: Secure credential setup - description: Exact models and provider routes available through connected accounts. name: Provider models - description: Create, label, list, and revoke organization API keys. name: API keys - description: Organization members and invitations. name: Members - description: Metered usage and quota state. name: Usage - description: Account, plan and allowance snapshot. name: Account - description: Checkout, billing portal and billing profile. name: Billing - description: Webhook endpoints, deliveries and replay. name: Webhooks - description: Organization-scoped compliance export. name: Compliance export - description: >- Models, agents, and skills Millwork can choose for a run. The API calls each one an arm. name: Registered options - description: >- Checks that score output and can stop, retry, or choose another option within the request policy. name: Output checks - description: Routing proposals and their review lifecycle. name: Repair proposals - description: Aggregate evaluation trend and repair history. name: Evaluation - description: Publisher and release identity records for supported models. name: Model definitions - description: >- Protected provider-credential references. Secret material never appears in this API. name: Credential references - description: Dashboard browser sign-in, sign-up, session lookup, and sign-out. name: Browser access - description: Read-only demo session setup. name: Demo - description: Machine setup and the authenticated organization profile. name: Organizations - description: Eligible immutable starter presets and typed blockers. name: Tenant templates - description: Stateless starter plan resolution. No write and no provider call. name: Tenant template plans - description: Durable starter applications that resume instead of restarting. name: Tenant template applications paths: /v1/tenant-template-applications: post: tags: - Tenant template applications summary: Apply tenant template description: Create or resume a tenant-template application. operationId: postTenantTemplateApplications parameters: - in: header name: Idempotency-Key required: true schema: minLength: 1 type: string requestBody: content: application/json: schema: additionalProperties: false properties: application_key: maxLength: 512 minLength: 1 type: string digest: minLength: 8 type: string issued_at: minLength: 1 type: string model_deployment_id: minLength: 1 type: string template_id: enum: - starter - pooled-open-model - byok-open-model type: string write: default: false type: boolean required: - digest - issued_at type: object required: true responses: '200': content: application/json: schema: additionalProperties: false properties: application_id: type: string completed_effects: items: additionalProperties: false properties: at: format: date-time type: string id: type: string required: - id - at type: object type: array consent: anyOf: - additionalProperties: false properties: attempt: minimum: 1 type: integer continue_url: format: uri type: string expires_at: format: date-time type: string handoff_intent_id: type: string required: - handoff_intent_id - continue_url - expires_at - attempt type: object - type: 'null' created_at: format: date-time type: string diagnostics: additionalProperties: true type: object echo_execution_id: anyOf: - type: string - type: 'null' echo_receipt_id: anyOf: - type: string - type: 'null' live_execution_id: anyOf: - type: string - type: 'null' live_proof: anyOf: - additionalProperties: false properties: digest: pattern: ^sha256:[0-9a-f]{64}$ type: string expires_at: format: date-time type: string issued_at: format: date-time type: string required: - digest - issued_at - expires_at type: object - type: 'null' managed_arm_id: anyOf: - type: string - type: 'null' next_action: additionalProperties: false properties: detail: type: string type: type: string required: - type - detail type: object receipt: anyOf: - additionalProperties: false properties: href: type: string receipt_id: type: string required: - receipt_id - href type: object - type: 'null' request_preset_id: anyOf: - type: string - type: 'null' result: anyOf: - additionalProperties: false properties: execution_id: type: string href: type: string model_provenance: additionalProperties: false properties: deployment: additionalProperties: false properties: fallback_policy: enum: - none - source_controlled type: string location: type: string model_deployment_id: type: string pricing_basis: enum: - source_reported - estimated type: string protocol_profile: enum: - openai_responses - openai_chat - anthropic_messages - gemini_generate_content - bedrock_converse type: string route: type: string required: - model_deployment_id - protocol_profile - location - route - pricing_basis - fallback_policy type: object requested: additionalProperties: false properties: family: type: string model_key: type: string publisher_id: type: string release: type: string served_variant_id: type: string upstream_ref: type: string required: - model_key - publisher_id - family - release - served_variant_id - upstream_ref type: object resolved: anyOf: - additionalProperties: false properties: attestation: enum: - source_response - gateway_response type: string source_request_id: anyOf: - type: string - type: 'null' upstream_ref: type: string required: - upstream_ref - source_request_id - attestation type: object - type: 'null' schema_version: enum: - model-attempt-provenance/v1 type: string source: additionalProperties: false properties: access_lane: enum: - byok - millwork_pool type: string auth_scheme: enum: - api_key - oauth2 - aws_sts_sigv4 type: string commercial_owner: enum: - customer - millwork type: string connection_id: type: string source_id: type: string source_kind: enum: - publisher_direct - gateway - cloud - fixture type: string required: - source_id - source_kind - connection_id - access_lane - commercial_owner - auth_scheme type: object required: - schema_version - requested - source - deployment - resolved type: object required: - execution_id - href - model_provenance type: object - type: 'null' selected_model_deployment_id: anyOf: - type: string - type: 'null' source_connection_id: anyOf: - type: string - type: 'null' state: enum: - planned - applying - consent_pending - arm_ready - echo_proved - live_queued - live_running - ready - action_required - failed_safe type: string template_id: enum: - starter - pooled-open-model - byok-open-model type: string template_version: type: string required: - application_id - template_id - template_version - state - completed_effects - result - receipt - managed_arm_id - echo_execution_id - echo_receipt_id - live_execution_id - selected_model_deployment_id - source_connection_id - consent - live_proof - next_action - diagnostics - created_at type: object description: Default Response 4XX: content: application/problem+json: schema: $ref: '#/components/schemas/Problem' description: >- Request failure in RFC 7807 format. The `type` field identifies the reason. Validation errors list affected fields. Rate and quota errors include `retry_after_s` and the `Retry-After` header. 5XX: content: application/problem+json: schema: $ref: '#/components/schemas/Problem' description: >- Server failure in RFC 7807 format. The response does not include internal error details. security: - bearerAuth: [] - sessionCookie: [] components: schemas: Problem: additionalProperties: false properties: detail: description: Short, safe, occurrence-specific description. Never a stack trace. type: string errors: description: Per-field validation failures (validation_failed only). items: additionalProperties: false properties: field: type: string message: type: string required: - field - message type: object type: array instance: description: Unique id for this occurrence, safe to quote in support requests. format: uuid type: string retry_after_s: description: >- Seconds to wait before retrying (rate_limited and quota_exceeded only; mirrored as the Retry-After header). type: integer status: description: HTTP status code, duplicated in the body per RFC 7807. type: integer title: description: Human-readable summary of the problem type. type: string type: description: Stable URI that identifies the error type. format: uri type: string required: - type - title - status - instance type: object securitySchemes: bearerAuth: description: Organization API key for application requests. scheme: bearer type: http sessionCookie: description: Dashboard browser session. Do not use this credential in an application. in: cookie name: solverapi_session type: apiKey ````