# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: 'API Evangelist conversational phrasing for Microsoft Azure Files: API'
  version: 1.0.0
extends: openapi/microsoft-azure-files-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-09-26'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 27
- target: $.paths['/files'].get
  update:
    x-apievangelist-phrasing:
      intent: List files owned by an Azure OpenAI resource
      effect: read
      questions:
      - Which training and validation files have I uploaded to my Azure OpenAI resource?
      - Can I see every file my OpenAI resource owns, including fine-tune result files?
      instructions:
      - text: List all files on my Azure OpenAI resource using API version {api-version}.
        slots:
          api-version: query.api-version
      - text: Show me the fine-tune files currently stored on the OpenAI resource.
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files'].post
  update:
    x-apievangelist-phrasing:
      intent: Upload a local file for fine-tuning
      effect: write
      questions:
      - How do I upload a JSONL training file from my machine to Azure OpenAI?
      - Can I push a local validation dataset to be used by a fine-tune job?
      instructions:
      - text: Upload {file} to Azure OpenAI with purpose {purpose}.
        slots:
          file: requestBody.file
          purpose: requestBody.purpose
      - text: Send local file {file} up as a fine-tune training file ({purpose}).
        slots:
          file: requestBody.file
          purpose: requestBody.purpose
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{file_id}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get an OpenAI file's status and size (file_id route)
      effect: read
      questions:
      - What is the processing status, size and purpose of one uploaded OpenAI file, looked up by file_id?
      - Has my fine-tune training file finished importing, using the underscore file_id route?
      instructions:
      - text: Get details for OpenAI file {file_id} via the file_id route.
        slots:
          file_id: path.file_id
      - text: Check status and byte size of training file {file_id} using the underscore file_id path.
        slots:
          file_id: path.file_id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{file_id}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete an OpenAI file (file_id route)
      effect: destructive
      questions:
      - Can I remove an uploaded file by file_id even after a fine-tune job trained on it?
      - What happens if I delete a training file that was used by a finished fine-tune, via the file_id route?
      instructions:
      - text: Delete OpenAI file {file_id} using the underscore file_id path.
        slots:
          file_id: path.file_id
      - text: Remove old training upload {file_id} through the file_id endpoint.
        slots:
          file_id: path.file_id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{file_id}/content'].get
  update:
    x-apievangelist-phrasing:
      intent: Download an OpenAI file's content (file_id route)
      effect: read
      questions:
      - Can I download the result metrics a fine-tune job produced, using the file_id content route?
      - Where do I fetch the raw bytes of an uploaded file by its underscore file_id?
      instructions:
      - text: Download the content of file {file_id} from the file_id content route.
        slots:
          file_id: path.file_id
      - text: Fetch the fine-tune results file {file_id} body via /files/{file_id}/content.
        slots:
          file_id: path.file_id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/import'].post
  update:
    x-apievangelist-phrasing:
      intent: Import a file into Azure OpenAI from a URL
      effect: write
      questions:
      - Can I bring a training file into Azure OpenAI straight from a blob URL instead of uploading it?
      - How do I import a fine-tune dataset hosted at a remote URL?
      instructions:
      - text: Import {content_url} as {filename} with purpose {purpose}.
        slots:
          content_url: requestBody.content_url
          filename: requestBody.filename
          purpose: requestBody.purpose
      - text: Pull the dataset at URL {content_url} into OpenAI and name it {filename} for {purpose}.
        slots:
          content_url: requestBody.content_url
          filename: requestBody.filename
          purpose: requestBody.purpose
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{file-id}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get an OpenAI file's details (file-id route)
      effect: read
      questions:
      - Which purpose and size does an uploaded file report on the hyphenated file-id path?
      - Is a given file still processing when I query it through the hyphenated file-id path?
      instructions:
      - text: Look up file metadata for {file-id} on the hyphenated file-id route.
        slots:
          file-id: path.file-id
      - text: Read status, purpose and size of {file-id} through the file-id path variant.
        slots:
          file-id: path.file-id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{file-id}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete an OpenAI file (file-id route)
      effect: destructive
      questions:
      - Can I erase a file through the hyphenated file-id delete path?
      - Does deleting via the hyphenated file-id route still work for files a fine-tune consumed?
      instructions:
      - text: Erase file {file-id} using the hyphenated file-id delete route.
        slots:
          file-id: path.file-id
      - text: Permanently drop upload {file-id} via the file-id variant.
        slots:
          file-id: path.file-id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{file-id}/content'].get
  update:
    x-apievangelist-phrasing:
      intent: Download an OpenAI file's content (file-id route)
      effect: read
      questions:
      - Can I read back what a service-generated file contains via the hyphenated file-id content path?
      - Which route returns file bytes when my client uses file-id rather than file_id?
      instructions:
      - text: Stream the bytes of file {file-id} from the file-id content endpoint.
        slots:
          file-id: path.file-id
      - text: Save the contents of generated file {file-id} using the hyphenated route.
        slots:
          file-id: path.file-id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{fileId}'].get
  update:
    x-apievangelist-phrasing:
      intent: Retrieve file information by fileId
      effect: read
      questions:
      - Can I get information about a file by fileId without pulling down its contents?
      - What metadata comes back from the camel-case fileId file lookup?
      instructions:
      - text: Retrieve file info for {fileId} without downloading its content.
        slots:
          fileId: path.fileId
      - text: Show the record for camel-case fileId {fileId}.
        slots:
          fileId: path.fileId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/files/{fileId}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a previously uploaded file by fileId
      effect: destructive
      questions:
      - Can I delete a previously uploaded file addressed by camel-case fileId?
      - Is there a delete for files on the camel-case fileId path?
      instructions:
      - text: Delete previously uploaded file {fileId}.
        slots:
          fileId: path.fileId
      - text: Get rid of the upload with camel-case fileId {fileId}.
        slots:
          fileId: path.fileId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/jobs/{jobId}/tasks/{taskId}/files/{filePath}'].get
  update:
    x-apievangelist-phrasing:
      intent: Download a file a Batch task wrote
      effect: read
      questions:
      - How do I read the stdout.txt a Batch task produced on its compute node?
      - Can I fetch only a byte range of a task's output file?
      instructions:
      - text: Download {filePath} from task {taskId} in Batch job {jobId}.
        slots:
          filePath: path.filePath
          taskId: path.taskId
          jobId: path.jobId
      - text: Get bytes {ocp-range} of task file {filePath} for task {taskId} in job {jobId}.
        slots:
          ocp-range: header.ocp-range
          filePath: path.filePath
          taskId: path.taskId
          jobId: path.jobId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/jobs/{jobId}/tasks/{taskId}/files/{filePath}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a Batch task's file from its node
      effect: destructive
      questions:
      - Can I clean up an output file a task left on the compute node where it ran?
      - Is it possible to recursively delete a directory in a task's working folder?
      instructions:
      - text: Delete task file {filePath} for task {taskId} in job {jobId}.
        slots:
          filePath: path.filePath
          taskId: path.taskId
          jobId: path.jobId
      - text: Recursively remove directory {filePath} from task {taskId} of job {jobId} (recursive={recursive}).
        slots:
          filePath: path.filePath
          taskId: path.taskId
          jobId: path.jobId
          recursive: query.recursive
      method: generated
      generated: '2026-09-26'
- target: $.paths['/jobs/{jobId}/tasks/{taskId}/files/{filePath}'].head
  update:
    x-apievangelist-phrasing:
      intent: Get properties of a Batch task file
      effect: read
      questions:
      - What size and last-modified time does a task's output file have, without downloading it?
      - Can I check whether a task file changed since a given date using only its headers?
      instructions:
      - text: Get the properties of {filePath} for task {taskId} in job {jobId}.
        slots:
          filePath: path.filePath
          taskId: path.taskId
          jobId: path.jobId
      - text: Check if task file {filePath} on task {taskId}, job {jobId} was modified since {If-Modified-Since}.
        slots:
          filePath: path.filePath
          taskId: path.taskId
          jobId: path.jobId
          If-Modified-Since: header.If-Modified-Since
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pools/{poolId}/nodes/{nodeId}/files/{filePath}'].get
  update:
    x-apievangelist-phrasing:
      intent: Download a file from a Batch compute node
      effect: read
      questions:
      - How do I pull a log file straight off a compute node in a Batch pool?
      - Can I read part of a node file by byte range rather than the whole thing?
      instructions:
      - text: Download {filePath} from node {nodeId} in pool {poolId}.
        slots:
          filePath: path.filePath
          nodeId: path.nodeId
          poolId: path.poolId
      - text: Read range {ocp-range} of node file {filePath} on compute node {nodeId}, pool {poolId}.
        slots:
          ocp-range: header.ocp-range
          filePath: path.filePath
          nodeId: path.nodeId
          poolId: path.poolId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pools/{poolId}/nodes/{nodeId}/files/{filePath}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a file from a Batch compute node
      effect: destructive
      questions:
      - Can I free disk space by deleting a file directly on a pool node?
      - Does removing a folder on a compute node support a recursive flag?
      instructions:
      - text: Delete {filePath} from compute node {nodeId} in pool {poolId}.
        slots:
          filePath: path.filePath
          nodeId: path.nodeId
          poolId: path.poolId
      - text: Wipe folder {filePath} on node {nodeId} of pool {poolId} with recursive set to {recursive}.
        slots:
          filePath: path.filePath
          nodeId: path.nodeId
          poolId: path.poolId
          recursive: query.recursive
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pools/{poolId}/nodes/{nodeId}/files/{filePath}'].head
  update:
    x-apievangelist-phrasing:
      intent: Get properties of a file on a compute node
      effect: read
      questions:
      - What is the size and timestamp of a file sitting on a Batch pool node?
      - Can I tell whether a node file is unchanged since a date without fetching it?
      instructions:
      - text: Get the properties of node file {filePath} on {nodeId} in pool {poolId}.
        slots:
          filePath: path.filePath
          nodeId: path.nodeId
          poolId: path.poolId
      - text: Check node {nodeId} (pool {poolId}) file {filePath} headers only, unmodified since {If-Unmodified-Since}.
        slots:
          nodeId: path.nodeId
          poolId: path.poolId
          filePath: path.filePath
          If-Unmodified-Since: header.If-Unmodified-Since
      method: generated
      generated: '2026-09-26'
- target: $.paths['/jobs/{jobId}/tasks/{taskId}/files'].get
  update:
    x-apievangelist-phrasing:
      intent: List the files in a Batch task's directory
      effect: read
      questions:
      - Which files did a Batch task write into its working directory?
      - Can I list a task's files recursively and filter by name?
      instructions:
      - text: List files for task {taskId} in job {jobId}.
        slots:
          taskId: path.taskId
          jobId: path.jobId
      - text: Recursively list task {taskId} files in job {jobId} matching {filter}.
        slots:
          taskId: path.taskId
          jobId: path.jobId
          filter: query.$filter
      method: generated
      generated: '2026-09-26'
- target: $.paths['/pools/{poolId}/nodes/{nodeId}/files'].get
  update:
    x-apievangelist-phrasing:
      intent: List task directory files on a compute node
      effect: read
      questions:
      - What files exist across all task directories on one Batch node?
      - Can I cap how many node files come back in one listing?
      instructions:
      - text: List all files on compute node {nodeId} in pool {poolId}.
        slots:
          nodeId: path.nodeId
          poolId: path.poolId
      - text: Show up to {maxresults} files from node {nodeId} of pool {poolId}.
        slots:
          maxresults: query.maxresults
          nodeId: path.nodeId
          poolId: path.poolId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/subscriptions/{subscriptionId}/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files'].get
  update:
    x-apievangelist-phrasing:
      intent: List files in a support file workspace
      effect: read
      questions:
      - Which files have I attached to a support file workspace in my subscription?
      - Can I see everything uploaded to a support ticket workspace scoped to one subscription?
      instructions:
      - text: List files in support workspace {fileWorkspaceName} under subscription {subscriptionId}.
        slots:
          fileWorkspaceName: path.fileWorkspaceName
          subscriptionId: path.subscriptionId
      - text: Show the attachments in subscription {subscriptionId}'s support workspace {fileWorkspaceName}.
        slots:
          subscriptionId: path.subscriptionId
          fileWorkspaceName: path.fileWorkspaceName
      method: generated
      generated: '2026-09-26'
- target: $.paths['/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files'].get
  update:
    x-apievangelist-phrasing:
      intent: List support workspace files without a subscription
      effect: read
      questions:
      - Can I list the files in a support workspace when I have no subscription to scope it to?
      - What attachments are in a tenant-level support file workspace?
      instructions:
      - text: List files in tenant-level support workspace {fileWorkspaceName}.
        slots:
          fileWorkspaceName: path.fileWorkspaceName
      - text: Show attachments in no-subscription workspace {fileWorkspaceName}.
        slots:
          fileWorkspaceName: path.fileWorkspaceName
      method: generated
      generated: '2026-09-26'
- target: $.paths['/subscriptions/{subscriptionId}/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files/{fileName}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a support workspace file in a subscription
      effect: read
      questions:
      - What are the details of one file in my subscription's support workspace?
      - Can I check the size of a support attachment I uploaded under a subscription?
      instructions:
      - text: Get support file {fileName} in workspace {fileWorkspaceName} for subscription {subscriptionId}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
          subscriptionId: path.subscriptionId
      - text: Show details of attachment {fileName} from subscription {subscriptionId} workspace {fileWorkspaceName}.
        slots:
          fileName: path.fileName
          subscriptionId: path.subscriptionId
          fileWorkspaceName: path.fileWorkspaceName
      method: generated
      generated: '2026-09-26'
- target: $.paths['/subscriptions/{subscriptionId}/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files/{fileName}'].put
  update:
    x-apievangelist-phrasing:
      intent: Create a support file in a subscription workspace
      effect: write
      questions:
      - How do I register a new file in a support workspace before uploading its content?
      - Can I create a support attachment entry scoped to my subscription?
      instructions:
      - text: Create file {fileName} in support workspace {fileWorkspaceName} for subscription {subscriptionId}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
          subscriptionId: path.subscriptionId
      - text: Register attachment {fileName} under subscription {subscriptionId} workspace {fileWorkspaceName} with properties {properties}.
        slots:
          fileName: path.fileName
          subscriptionId: path.subscriptionId
          fileWorkspaceName: path.fileWorkspaceName
          properties: requestBody.properties
      method: generated
      generated: '2026-09-26'
- target: $.paths['/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files/{fileName}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a support file without a subscription
      effect: read
      questions:
      - What does a support attachment look like in a workspace that isn't tied to a subscription?
      - Can I look up one file in a tenant-level support workspace?
      instructions:
      - text: Get file {fileName} from tenant-level support workspace {fileWorkspaceName}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
      - text: Look up no-subscription attachment {fileName} in {fileWorkspaceName}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
      method: generated
      generated: '2026-09-26'
- target: $.paths['/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files/{fileName}'].put
  update:
    x-apievangelist-phrasing:
      intent: Create a support file without a subscription
      effect: write
      questions:
      - Can I create a support attachment in a workspace when I have no Azure subscription?
      - Which call adds a new file entry to a tenant-level support workspace?
      instructions:
      - text: Create file {fileName} in tenant-level support workspace {fileWorkspaceName}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
      - text: Add no-subscription attachment {fileName} to {fileWorkspaceName} with properties {properties}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
          properties: requestBody.properties
      method: generated
      generated: '2026-09-26'
- target: $.paths['/subscriptions/{subscriptionId}/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files/{fileName}/upload'].post
  update:
    x-apievangelist-phrasing:
      intent: Upload content to a support file in a subscription
      effect: write
      questions:
      - How do I upload the bytes of a support attachment in chunks under my subscription?
      - Can I send file content to a support workspace one chunk index at a time?
      instructions:
      - text: Upload content to support file {fileName} in {fileWorkspaceName} for subscription {subscriptionId}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
          subscriptionId: path.subscriptionId
      - text: Send chunk {chunkIndex} of {fileName} to workspace {fileWorkspaceName} in subscription {subscriptionId}.
        slots:
          chunkIndex: requestBody.chunkIndex
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
          subscriptionId: path.subscriptionId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/providers/Microsoft.Support/fileWorkspaces/{fileWorkspaceName}/files/{fileName}/upload'].post
  update:
    x-apievangelist-phrasing:
      intent: Upload content to a support file without a subscription
      effect: write
      questions:
      - Can I upload attachment content to a tenant-level support workspace with no subscription?
      - Which endpoint accepts chunked content for a no-subscription support file?
      instructions:
      - text: Upload content to tenant-level support file {fileName} in {fileWorkspaceName}.
        slots:
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
      - text: Push chunk number {chunkIndex} of no-subscription file {fileName} into {fileWorkspaceName}.
        slots:
          chunkIndex: requestBody.chunkIndex
          fileName: path.fileName
          fileWorkspaceName: path.fileWorkspaceName
      method: generated
      generated: '2026-09-26'