Microsoft Azure Batch · Schema
BatchTaskGroup
A collection of Azure Batch Tasks to add.
BatchComputeJob SchedulingHigh Performance ComputingCloudMicrosoftMicrosoft AzureParallel ProcessingSchedulingInfrastructure
Properties
| Name | Type | Description |
|---|---|---|
| value | array | The collection of Tasks to add. The maximum count of Tasks is 100. The total serialized size of this collection must be less than 1MB. If it is greater than 1MB (for example if each Task has 100's of |
JSON Schema
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "https://artifacts.apis.io/microsoft-azure-batch/json-schema/microsoft-azure-batch-batch-task-group-schema.json",
"title": "BatchTaskGroup",
"description": "A collection of Azure Batch Tasks to add.",
"x-generated": "2026-11-05",
"x-method": "derived",
"x-generator": "derive-json-schema.py",
"x-source": "openapi/_superseded/microsoft-azure-batch-tasks-api-openapi.yml#/components/schemas/BatchTaskGroup",
"type": "object",
"properties": {
"value": {
"type": "array",
"description": "The collection of Tasks to add. The maximum count of Tasks is 100. The total serialized size of this collection must be less than 1MB. If it is greater than 1MB (for example if each Task has 100's of resource files or environment variables), the request will fail with code 'RequestBodyTooLarge' and should be retried again with fewer Tasks.",
"items": {
"$ref": "#/$defs/BatchTaskCreateOptions"
},
"x-ms-client-name": "values"
}
},
"required": [
"value"
],
"$defs": {
"AutoUserScope": {
"type": "string",
"description": "AutoUserScope enums",
"enum": [
"task",
"pool"
],
"x-ms-enum": {
"name": "AutoUserScope",
"modelAsString": true,
"values": [
{
"name": "Task",
"value": "task",
"description": "Specifies that the service should create a new user for the Task."
},
{
"name": "Pool",
"value": "pool",
"description": "Specifies that the Task runs as the common auto user Account which is created on every Compute Node in a Pool."
}
]
}
},
"AutoUserSpecification": {
"type": "object",
"description": "Specifies the options for the auto user that runs an Azure Batch Task.",
"properties": {
"scope": {
"$ref": "#/$defs/AutoUserScope",
"description": "The scope for the auto user. The default value is pool. If the pool is running Windows a value of Task should be specified if stricter isolation between tasks is required. For example, if the task mutates the registry in a way which could impact other tasks."
},
"elevationLevel": {
"$ref": "#/$defs/ElevationLevel",
"description": "The elevation level of the auto user. The default value is nonAdmin."
}
}
},
"BatchAffinityInfo": {
"type": "object",
"description": "A locality hint that can be used by the Batch service to select a Compute Node\non which to start a Task.",
"properties": {
"affinityId": {
"type": "string",
"description": "An opaque string representing the location of a Compute Node or a Task that has run previously. You can pass the affinityId of a Node to indicate that this Task needs to run on that Compute Node. Note that this is just a soft affinity. If the target Compute Node is busy or unavailable at the time the Task is scheduled, then the Task will be scheduled elsewhere."
}
},
"required": [
"affinityId"
]
},
"BatchApplicationPackageReference": {
"type": "object",
"description": "A reference to an Package to be deployed to Compute Nodes.",
"properties": {
"applicationId": {
"type": "string",
"description": "The ID of the application to deploy. When creating a pool, the package's application ID must be fully qualified (/subscriptions/{subscriptionId}/resourceGroups/{resourceGroupName}/providers/Microsoft.Batch/batchAccounts/{accountName}/applications/{applicationName})."
},
"version": {
"type": "string",
"description": "The version of the application to deploy. If omitted, the default version is deployed. If this is omitted on a Pool, and no default version is specified for this application, the request fails with the error code InvalidApplicationPackageReferences and HTTP status code 409. If this is omitted on a Task, and no default version is specified for this application, the Task fails with a pre-processing error."
}
},
"required": [
"applicationId"
]
},
"BatchJobActionKind": {
"type": "string",
"description": "BatchJobAction enums",
"enum": [
"none",
"disable",
"terminate"
],
"x-ms-enum": {
"name": "BatchJobActionKind",
"modelAsString": true,
"values": [
{
"name": "None",
"value": "none",
"description": "Take no action."
},
{
"name": "Disable",
"value": "disable",
"description": "Disable the Job. This is equivalent to calling the disable Job API, with a disableTasks value of requeue."
},
{
"name": "Terminate",
"value": "terminate",
"description": "Terminate the Job. The terminationReason in the Job's executionInfo is set to \"TaskFailed\"."
}
]
}
},
"BatchNodeIdentityReference": {
"type": "object",
"description": "The reference to a user assigned identity associated with the Batch pool which\na compute node will use.",
"properties": {
"resourceId": {
"type": "string",
"format": "arm-id",
"description": "The ARM resource id of the user assigned identity."
}
}
},
"BatchTaskConstraints": {
"type": "object",
"description": "Execution constraints to apply to a Task.",
"properties": {
"maxWallClockTime": {
"type": "string",
"format": "duration",
"description": "The maximum elapsed time that the Task may run, measured from the time the Task starts. If the Task does not complete within the time limit, the Batch service terminates it. If this is not specified, there is no time limit on how long the Task may run. The time duration is specified in ISO 8601 format."
},
"retentionTime": {
"type": "string",
"format": "duration",
"description": "The minimum time to retain the Task directory on the Compute Node where it ran, from the time it completes execution. After this time, the Batch service may delete the Task directory and all its contents. The default is 7 days, i.e. the Task directory will be retained for 7 days unless the Compute Node is removed or the Job is deleted. The time duration is specified in ISO 8601 format."
},
"maxTaskRetryCount": {
"type": "integer",
"format": "int32",
"description": "The maximum number of times the Task may be retried. The Batch service retries a Task if its exit code is nonzero. Note that this value specifically controls the number of retries for the Task executable due to a nonzero exit code. The Batch service will try the Task once, and may then retry up to this limit. For example, if the maximum retry count is 3, Batch tries the Task up to 4 times (one initial try and 3 retries). If the maximum retry count is 0, the Batch service does not retry the Task after the first attempt. If the maximum retry count is -1, the Batch service retries the Task without limit, however this is not recommended for a start task or any task. The default value is 0 (no retries)."
}
}
},
"BatchTaskContainerSettings": {
"type": "object",
"description": "The container settings for a Task.",
"properties": {
"containerRunOptions": {
"type": "string",
"description": "Additional options to the container create command. These additional options are supplied as arguments to the \"docker create\" command, in addition to those controlled by the Batch Service."
},
"imageName": {
"type": "string",
"description": "The Image to use to create the container in which the Task will run. This is the full Image reference, as would be specified to \"docker pull\". If no tag is provided as part of the Image name, the tag \":latest\" is used as a default."
},
"registry": {
"$ref": "#/$defs/ContainerRegistryReference",
"description": "The private registry which contains the container Image. This setting can be omitted if was already provided at Pool creation."
},
"workingDirectory": {
"$ref": "#/$defs/ContainerWorkingDirectory",
"description": "The location of the container Task working directory. The default is 'taskWorkingDirectory'."
},
"containerHostBatchBindMounts": {
"type": "array",
"description": "The paths you want to mounted to container task. If this array is null or be not present, container task will mount entire temporary disk drive in windows (or AZ_BATCH_NODE_ROOT_DIR in Linux). It won't' mount any data paths into container if this array is set as empty.",
"items": {
"$ref": "#/$defs/ContainerHostBatchBindMountEntry"
}
}
},
"required": [
"imageName"
]
},
"BatchTaskCreateOptions": {
"type": "object",
"description": "Parameters for creating an Azure Batch Task.",
"properties": {
"id": {
"type": "string",
"description": "A string that uniquely identifies the Task within the Job. The ID can contain any combination of alphanumeric characters including hyphens and underscores, and cannot contain more than 64 characters. The ID is case-preserving and case-insensitive (that is, you may not have two IDs within a Job that differ only by case)."
},
"displayName": {
"type": "string",
"description": "A display name for the Task. The display name need not be unique and can contain any Unicode characters up to a maximum length of 1024."
},
"exitConditions": {
"$ref": "#/$defs/ExitConditions",
"description": "How the Batch service should respond when the Task completes."
},
"commandLine": {
"type": "string",
"description": "The command line of the Task. For multi-instance Tasks, the command line is executed as the primary Task, after the primary Task and all subtasks have finished executing the coordination command line. The command line does not run under a shell, and therefore cannot take advantage of shell features such as environment variable expansion. If you want to take advantage of such features, you should invoke the shell in the command line, for example using \"cmd /c MyCommand\" in Windows or \"/bin/sh -c MyCommand\" in Linux. If the command line refers to file paths, it should use a relative path (relative to the Task working directory), or use the Batch provided environment variable (https://learn.microsoft.com/azure/batch/batch-compute-node-environment-variables)."
},
"containerSettings": {
"$ref": "#/$defs/BatchTaskContainerSettings",
"description": "The settings for the container under which the Task runs. If the Pool that will run this Task has containerConfiguration set, this must be set as well. If the Pool that will run this Task doesn't have containerConfiguration set, this must not be set. When this is specified, all directories recursively below the AZ_BATCH_NODE_ROOT_DIR (the root of Azure Batch directories on the node) are mapped into the container, all Task environment variables are mapped into the container, and the Task command line is executed in the container. Files produced in the container outside of AZ_BATCH_NODE_ROOT_DIR might not be reflected to the host disk, meaning that Batch file APIs will not be able to access those files."
},
"resourceFiles": {
"type": "array",
"description": "A list of files that the Batch service will download to the Compute Node before running the command line. For multi-instance Tasks, the resource files will only be downloaded to the Compute Node on which the primary Task is executed. There is a maximum size for the list of resource files. When the max size is exceeded, the request will fail and the response error code will be RequestEntityTooLarge. If this occurs, the collection of ResourceFiles must be reduced in size. This can be achieved using .zip files, Application Packages, or Docker Containers.",
"items": {
"$ref": "#/$defs/ResourceFile"
}
},
"outputFiles": {
"type": "array",
"description": "A list of files that the Batch service will upload from the Compute Node after running the command line. For multi-instance Tasks, the files will only be uploaded from the Compute Node on which the primary Task is executed.",
"items": {
"$ref": "#/$defs/OutputFile"
}
},
"environmentSettings": {
"type": "array",
"description": "A list of environment variable settings for the Task.",
"items": {
"$ref": "#/$defs/EnvironmentSetting"
}
},
"affinityInfo": {
"$ref": "#/$defs/BatchAffinityInfo",
"description": "A locality hint that can be used by the Batch service to select a Compute Node on which to start the new Task."
},
"constraints": {
"$ref": "#/$defs/BatchTaskConstraints",
"description": "The execution constraints that apply to this Task. If you do not specify constraints, the maxTaskRetryCount is the maxTaskRetryCount specified for the Job, the maxWallClockTime is infinite, and the retentionTime is 7 days."
},
"requiredSlots": {
"type": "integer",
"format": "int32",
"description": "The number of scheduling slots that the Task required to run. The default is 1. A Task can only be scheduled to run on a compute node if the node has enough free scheduling slots available. For multi-instance Tasks, this must be 1."
},
"userIdentity": {
"$ref": "#/$defs/UserIdentity",
"description": "The user identity under which the Task runs. If omitted, the Task runs as a non-administrative user unique to the Task."
},
"multiInstanceSettings": {
"$ref": "#/$defs/MultiInstanceSettings",
"description": "An object that indicates that the Task is a multi-instance Task, and contains information about how to run the multi-instance Task."
},
"dependsOn": {
"$ref": "#/$defs/BatchTaskDependencies",
"description": "The Tasks that this Task depends on. This Task will not be scheduled until all Tasks that it depends on have completed successfully. If any of those Tasks fail and exhaust their retry counts, this Task will never be scheduled. If the Job does not have usesTaskDependencies set to true, and this element is present, the request fails with error code TaskDependenciesNotSpecifiedOnJob."
},
"applicationPackageReferences": {
"type": "array",
"description": "A list of Packages that the Batch service will deploy to the Compute Node before running the command line. Application packages are downloaded and deployed to a shared directory, not the Task working directory. Therefore, if a referenced package is already on the Node, and is up to date, then it is not re-downloaded; the existing copy on the Compute Node is used. If a referenced Package cannot be installed, for example because the package has been deleted or because download failed, the Task fails.",
"items": {
"$ref": "#/$defs/BatchApplicationPackageReference"
}
}
},
"required": [
"id",
"commandLine"
]
},
"BatchTaskDependencies": {
"type": "object",
"description": "Specifies any dependencies of a Task. Any Task that is explicitly specified or\nwithin a dependency range must complete before the dependant Task will be\nscheduled.",
"properties": {
"taskIds": {
"type": "array",
"description": "The list of Task IDs that this Task depends on. All Tasks in this list must complete successfully before the dependent Task can be scheduled. The taskIds collection is limited to 64000 characters total (i.e. the combined length of all Task IDs). If the taskIds collection exceeds the maximum length, the Add Task request fails with error code TaskDependencyListTooLong. In this case consider using Task ID ranges instead.",
"items": {
"type": "string"
}
},
"taskIdRanges": {
"type": "array",
"description": "The list of Task ID ranges that this Task depends on. All Tasks in all ranges must complete successfully before the dependent Task can be scheduled.",
"items": {
"$ref": "#/$defs/BatchTaskIdRange"
}
}
}
},
"BatchTaskIdRange": {
"type": "object",
"description": "The start and end of the range are inclusive. For example, if a range has start\n9 and end 12, then it represents Tasks '9', '10', '11' and '12'.",
"properties": {
"start": {
"type": "integer",
"format": "int32",
"description": "The first Task ID in the range."
},
"end": {
"type": "integer",
"format": "int32",
"description": "The last Task ID in the range."
}
},
"required": [
"start",
"end"
]
},
"ContainerHostBatchBindMountEntry": {
"type": "object",
"description": "The entry of path and mount mode you want to mount into task container.",
"properties": {
"source": {
"$ref": "#/$defs/ContainerHostDataPath",
"description": "The path which be mounted to container customer can select."
},
"isReadOnly": {
"type": "boolean",
"description": "Mount this source path as read-only mode or not. Default value is false (read/write mode). For Linux, if you mount this path as a read/write mode, this does not mean that all users in container have the read/write access for the path, it depends on the access in host VM. If this path is mounted read-only, all users within the container will not be able to modify the path."
}
}
},
"ContainerHostDataPath": {
"type": "string",
"description": "The paths which will be mounted to container task's container.",
"enum": [
"Shared",
"Startup",
"VfsMounts",
"Task",
"JobPrep",
"Applications"
],
"x-ms-enum": {
"name": "ContainerHostDataPath",
"modelAsString": true,
"values": [
{
"name": "Shared",
"value": "Shared",
"description": "The path for multi-instances task to shared their files."
},
{
"name": "Startup",
"value": "Startup",
"description": "The path for start task."
},
{
"name": "VfsMounts",
"value": "VfsMounts",
"description": "The path contains all virtual file systems are mounted on this node."
},
{
"name": "Task",
"value": "Task",
"description": "The task path."
},
{
"name": "JobPrep",
"value": "JobPrep",
"description": "The job-prep task path."
},
{
"name": "Applications",
"value": "Applications",
"description": "The applications path."
}
]
}
},
"ContainerRegistryReference": {
"type": "object",
"description": "A private container registry.",
"properties": {
"username": {
"type": "string",
"description": "The user name to log into the registry server."
},
"password": {
"type": "string",
"format": "password",
"description": "The password to log into the registry server.",
"x-ms-secret": true
},
"registryServer": {
"type": "string",
"format": "uri",
"description": "The registry URL. If omitted, the default is \"docker.io\"."
},
"identityReference": {
"$ref": "#/$defs/BatchNodeIdentityReference",
"description": "The reference to the user assigned identity to use to access an Azure Container Registry instead of username and password."
}
}
},
"ContainerWorkingDirectory": {
"type": "string",
"description": "ContainerWorkingDirectory enums",
"enum": [
"taskWorkingDirectory",
"containerImageDefault"
],
"x-ms-enum": {
"name": "ContainerWorkingDirectory",
"modelAsString": true,
"values": [
{
"name": "TaskWorkingDirectory",
"value": "taskWorkingDirectory",
"description": "Use the standard Batch service Task working directory, which will contain the Task Resource Files populated by Batch."
},
{
"name": "ContainerImageDefault",
"value": "containerImageDefault",
"description": "Use the working directory defined in the container Image. Beware that this directory will not contain the Resource Files downloaded by Batch."
}
]
}
},
"DependencyAction": {
"type": "string",
"description": "DependencyAction enums",
"enum": [
"satisfy",
"block"
],
"x-ms-enum": {
"name": "DependencyAction",
"modelAsString": true,
"values": [
{
"name": "Satisfy",
"value": "satisfy",
"description": "Satisfy tasks waiting on this task; once all dependencies are satisfied, the task will be scheduled to run."
},
{
"name": "Block",
"value": "block",
"description": "Blocks tasks waiting on this task, preventing them from being scheduled."
}
]
}
},
"ElevationLevel": {
"type": "string",
"description": "ElevationLevel enums",
"enum": [
"nonadmin",
"admin"
],
"x-ms-enum": {
"name": "ElevationLevel",
"modelAsString": true,
"values": [
{
"name": "NonAdmin",
"value": "nonadmin",
"description": "The user is a standard user without elevated access."
},
{
"name": "Admin",
"value": "admin",
"description": "The user is a user with elevated access and operates with full Administrator permissions."
}
]
}
},
"EnvironmentSetting": {
"type": "object",
"description": "An environment variable to be set on a Task process.",
"properties": {
"name": {
"type": "string",
"description": "The name of the environment variable."
},
"value": {
"type": "string",
"description": "The value of the environment variable."
}
},
"required": [
"name"
]
},
"ExitCodeMapping": {
"type": "object",
"description": "How the Batch service should respond if a Task exits with a particular exit\ncode.",
"properties": {
"code": {
"type": "integer",
"format": "int32",
"description": "A process exit code."
},
"exitOptions": {
"$ref": "#/$defs/ExitOptions",
"description": "How the Batch service should respond if the Task exits with this exit code."
}
},
"required": [
"code",
"exitOptions"
]
},
"ExitCodeRangeMapping": {
"type": "object",
"description": "A range of exit codes and how the Batch service should respond to exit codes\nwithin that range.",
"properties": {
"start": {
"type": "integer",
"format": "int32",
"description": "The first exit code in the range."
},
"end": {
"type": "integer",
"format": "int32",
"description": "The last exit code in the range."
},
"exitOptions": {
"$ref": "#/$defs/ExitOptions",
"description": "How the Batch service should respond if the Task exits with an exit code in the range start to end (inclusive)."
}
},
"required": [
"start",
"end",
"exitOptions"
]
},
"ExitConditions": {
"type": "object",
"description": "Specifies how the Batch service should respond when the Task completes.",
"properties": {
"exitCodes": {
"type": "array",
"description": "A list of individual Task exit codes and how the Batch service should respond to them.",
"items": {
"$ref": "#/$defs/ExitCodeMapping"
}
},
"exitCodeRanges": {
"type": "array",
"description": "A list of Task exit code ranges and how the Batch service should respond to them.",
"items": {
"$ref": "#/$defs/ExitCodeRangeMapping"
}
},
"preProcessingError": {
"$ref": "#/$defs/ExitOptions",
"description": "How the Batch service should respond if the Task fails to start due to an error."
},
"fileUploadError": {
"$ref": "#/$defs/ExitOptions",
"description": "How the Batch service should respond if a file upload error occurs. If the Task exited with an exit code that was specified via exitCodes or exitCodeRanges, and then encountered a file upload error, then the action specified by the exit code takes precedence."
},
"default": {
"$ref": "#/$defs/ExitOptions",
"description": "How the Batch service should respond if the Task fails with an exit condition not covered by any of the other properties. This value is used if the Task exits with any nonzero exit code not listed in the exitCodes or exitCodeRanges collection, with a pre-processing error if the preProcessingError property is not present, or with a file upload error if the fileUploadError property is not present. If you want non-default behavior on exit code 0, you must list it explicitly using the exitCodes or exitCodeRanges collection."
}
}
},
"ExitOptions": {
"type": "object",
"description": "Specifies how the Batch service responds to a particular exit condition.",
"properties": {
"jobAction": {
"$ref": "#/$defs/BatchJobActionKind",
"description": "An action to take on the Job containing the Task, if the Task completes with the given exit condition and the Job's onTaskFailed property is 'performExitOptionsJobAction'. The default is none for exit code 0 and terminate for all other exit conditions. If the Job's onTaskFailed property is noaction, then specifying this property returns an error and the add Task request fails with an invalid property value error; if you are calling the REST API directly, the HTTP status code is 400 (Bad Request)."
},
"dependencyAction": {
"$ref": "#/$defs/DependencyAction",
"description": "An action that the Batch service performs on Tasks that depend on this Task. Possible values are 'satisfy' (allowing dependent tasks to progress) and 'block' (dependent tasks continue to wait). Batch does not yet support cancellation of dependent tasks."
}
}
},
"MultiInstanceSettings": {
"type": "object",
"description": "Multi-instance Tasks are commonly used to support MPI Tasks. In the MPI case,\nif any of the subtasks fail (for example due to exiting with a non-zero exit\ncode) the entire multi-instance Task fails. The multi-instance Task is then\nterminated and retried, up to its retry limit.",
"properties": {
"numberOfInstances": {
"type": "integer",
"format": "int32",
"description": "The number of Compute Nodes required by the Task. If omitted, the default is 1."
},
"coordinationCommandLine": {
"type": "string",
"description": "The command line to run on all the Compute Nodes to enable them to coordinate when the primary runs the main Task command. A typical coordination command line launches a background service and verifies that the service is ready to process inter-node messages."
},
"commonResourceFiles": {
"type": "array",
"description": "A list of files that the Batch service will download before running the coordination command line. The difference between common resource files and Task resource files is that common resource files are downloaded for all subtasks including the primary, whereas Task resource files are downloaded only for the primary. Also note that these resource files are not downloaded to the Task working directory, but instead are downloaded to the Task root directory (one directory above the working directory). There is a maximum size for the list of resource files. When the max size is exceeded, the request will fail and the response error code will be RequestEntityTooLarge. If this occurs, the collection of ResourceFiles must be reduced in size. This can be achieved using .zip files, Application Packages, or Docker Containers.",
"items": {
"$ref": "#/$defs/ResourceFile"
}
}
},
"required": [
"coordinationCommandLine"
]
},
"OutputFile": {
"type": "object",
"description": "On every file uploads, Batch service writes two log files to the compute node, 'fileuploadout.txt' and 'fileuploaderr.txt'. These log files are used to learn more about a specific failure.",
"properties": {
"filePattern": {
"type": "string",
"description": "A pattern indicating which file(s) to upload. Both relative and absolute paths are supported. Relative paths are relative to the Task working directory. The following wildcards are supported: * matches 0 or more characters (for example pattern abc* would match abc or abcdef), ** matches any directory, ? matches any single character, [abc] matches one character in the brackets, and [a-c] matches one character in the range. Brackets can include a negation to match any character not specified (for example [!abc] matches any character but a, b, or c). If a file name starts with \".\" it is ignored by default but may be matched by specifying it explicitly (for example *.gif will not match .a.gif, but .*.gif will). A simple example: **\\*.txt matches any file that does not start in '.' and ends with .txt in the Task working directory or any subdirectory. If the filename contains a wildcard character it can be escaped using brackets (for example abc[*] would match a file named abc*). Note that both \\ and / are treated as directory separators on Windows, but only / is on Linux. Environment variables (%var% on Windows or $var on Linux) are expanded prior to the pattern being applied."
},
"destination": {
"$ref": "#/$defs/OutputFileDestination",
"description": "The destination for the output file(s)."
},
"uploadOptions": {
"$ref": "#/$defs/OutputFileUploadConfig",
"description": "Additional options for the upload operation, including under what conditions to perform the upload."
}
},
"required": [
"filePattern",
"destination",
"uploadOptions"
]
},
"OutputFileBlobContainerDestination": {
"type": "object",
"description": "Specifies a file upload destination within an Azure blob storage container.",
"properties": {
"path": {
"type": "string",
"description": "The destination blob or virtual directory within the Azure Storage container. If filePattern refers to a specific file (i.e. contains no wildcards), then path is the name of the blob to which to upload that file. If filePattern contains one or more wildcards (and therefore may match multiple files), then path is the name of the blob virtual directory (which is prepended to each blob name) to which to upload the file(s). If omitted, file(s) are uploaded to the root of the container
# --- truncated at 32 KB (39 KB total) ---
# Full source: https://artifacts.apis.io/microsoft-azure-batch/json-schema/microsoft-azure-batch-batch-task-group-schema.json
Work with this as data
Every JSON Schema here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for schemas
4 MCP tools reach this
find_json_schemasBrowse and filter every JSON Schema in the catalog.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.
Call it yourself
curl for this page
This JSON Schema
curl "https://apis.io/api/v1/json-schemas/microsoft-azure-batch-batch-task-group"
All schemas
curl "https://apis.io/api/v1/json-schemas?limit=25"Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.