Skip to content

v1.2.md

Latest commit

 

History

History
203 lines (141 loc) · 14.9 KB

v1.2.md

File metadata and controls

203 lines (141 loc) · 14.9 KB

ASOR API Documentation v1.2

asor

Version: v1.2

Base URL: https://{workday-host}/api/asor/v1/{tenantname}/agentDefinition

The Agent System of Record Service is used to create and maintain Agent Definitions in Workday.

📋 Navigation

📡 Endpoints

GET /agentDefinition/{id}

Summary: This operation returns the specific Agent Definition including Skills and certain Tools used in those Skills.

Description: This operation returns the specific Agent Definition including Skills and certain Tools used in those Skills. Each definition can contain many Skills, and in those Skills they may contain Tools based on the Invocation / Execution Type. Generally REST APIs will be listed here.

Secured by: Setup: Agents

Responses

Status Code Description Schema
200 Successful response. A successful response can return no matched data. agentCardAndTools
400 Invalid request. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
401 Invalid resource or operation. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
403 User has insufficient permissions. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
404 Resource not found. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
default An error occurred. ERROR_MODEL_REFERENCE

POST /agentDefinition

Summary: Based on the A2A Agent Card specification, this REST operation allows you to create an Agent Definition in Workday.

Description: Based on the A2A Agent Card specification, this REST operation allows you to create an Agent Definition in Workday, while also highlighting the specific Workday APIs/Resources that the Agent is allowed access to. It will update an existing instance IF one already exists with the same name, provider organization, and version.

Secured by: Development

Parameters

Parameter Location Type Required Description
agentCardAndTools body agentCardAndTools ✅ Yes The Agent Definition that contains Skills optionally identifying Workday resources used in those Skills.

Request Body

Content Type: application/json

Schema: agentCardAndTools

Responses

Status Code Description Schema
201 Resource created. agentCardAndTools
400 Invalid request. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
401 Invalid resource or operation. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
403 User has insufficient permissions. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
404 Resource not found. (https://community.workday.com/rest/error-messages) VALIDATION_ERROR_MODEL_REFERENCE
default An error occurred. ERROR_MODEL_REFERENCE

📊 Models

agentCardAndTools {#agentcardandtools}

Property Type Description Required
id string The ID is part of the return payload when an Agent is created, it represents the internal Workday ID for this Agent. ❌ No
name string Human readable name of the agent. ✅ Yes
description string A human-readable description of the agent. Used to assist users and other agents in understanding what the agent can do. ✅ Yes
url string A URL to the address the agent is hosted at. ✅ Yes
overview string A message to display when summarizing this Agent or when this Agent is initially invoked. ❌ No
iconUrl string A URL to an icon for the agent. ❌ No
provider provider The service provider of the agent. ✅ Yes
platform platform The Platform that this Agent is executing on. If no appropriate Platform is available use OTHER. ✅ Yes
version string The version of the agent - format is up to the provider. ✅ Yes
documentationUrl string A URL to documentation for the agent. ❌ No
externalAgentID string A referenceable ID that is understood by the partner system, useful for callbacks from Workday. ❌ No
externalTenantID string A referenceable ID to locate where this Agent lives in the partner system to match the Workday tenant. ❌ No
capabilities capabilities Optional capabilities supported by the agent. ✅ Yes
defaultInputModes arraymimeType The set of interaction modes that the agent supports across all skills. This can be overridden per-skill. Supported media types for input. ❌ No
defaultOutputModes arraymimeType Supported media types for output. ❌ No
skills arrayagentskill Skills are a unit of capability that an agent can perform. ✅ Yes
supportsAuthenticatedExtendedCard boolean true if the agent supports providing an extended agent card when the user is authenticated. Defaults to false if not specified. ❌ No
workdayConfig arrayagentSkillResource Identify Workday Resources to be used in Agent Skills. ❌ No

provider {#provider}

The service provider of the agent

Property Type Description Required
id string Agent Provider Reference ID = "Provider={Provider Name} ✅ Yes

platform {#platform}

The execution platform for this agent

Property Type Description Required
id string Agent Platform Reference ID = "Platform={Provider Name} ✅ Yes

capabilities {#capabilities}

Optional capabilities supported by the agent.

Property Type Description Required
pushNotifications boolean true if the agent can notify updates to client. ❌ No
streaming boolean true if the agent supports SSE. ❌ No
stateTransitionHistory boolean true if the agent exposes status change history for tasks. ❌ No

mimeType {#mimetype}

Property Type Description Required
type string Supported media type. ✅ Yes

agentSkill {#agentskill}

Property Type Description Required
id string Unique identifier for the agent's skill. ✅ Yes
name string Human readable name of the skill. ✅ Yes
description string Description of the skill - will be used by the client or a human as a hint to understand what the skill does. ✅ Yes
tags arrayagentSkillTag Set of tagwords describing classes of capabilities for this specific skill. ✅ Yes
inputModes arraymimeType The set of interaction modes that the skill supports (if different than the default). ❌ No
outputModes arraymimeType Supported media types for output. ❌ No

agentSkillTag {#agentskilltag}

Property Type Description Required
tag string Will be used by the client as a hint to understand how the skill can be used. ✅ Yes

agentSkillResource {#agentskillresource}

Property Type Description Required
skillId string Unique identifier for the agent's skill. ✅ Yes
executionMode executionMode Unique identifier for the agent's skill. ✅ Yes
workdayResources arrayworkdayResource List of Workday Resources used by this agents skill. ✅ Yes

executionMode {#executionMode}

Property Type Description Required
id enum[Ambient, Delegate] If this Skill requires/necessitates human invocation, it is "Delegate". If no human is required to invoke it would be classed as "Ambient". This deliniates the Grant Type necessary. Note, must be in format "Mode=Ambient" or "Mode=Delegate" ✅ Yes

workdayResource {#workdayResource}

Property Type Description Required
agent_resource agent_resource Workday Resource for this Tool if applicable. ❌ No
tool_name string The Operation that defines what this Tool can do - i.e. Endpoint or Task. ✅ Yes
description string Description of the Operation. ❌ No
tools arraytools List of related Workday Securable Items. For example, Data Sources for WQL queries. ❌ No

agent_resource {#agent_resource}

Property Type Description Required
id string Workday ID of the Workday Tool required for this Skill. ✅ Yes

tools {#tools}

Property Type Description Required
id string Workday ID of the Workday Securable Item related to this Agent Tool - i.e. CRF Fields or Data Sources etc. ✅ Yes

VALIDATION_ERROR_MODEL_REFERENCE {#validation_error_model_reference}

Property Type Description Required
error string A description of the error ✅ Yes
errors array An array of validation errors ❌ No

ERROR_MODEL_REFERENCE {#error_model_reference}

Property Type Description Required
error string A description of the error ✅ Yes
code string The code that corresponds to the error message. Use the error code to drive programmatic error-handling behavior. Don't use error message strings for this purpose because they are subject to change ❌ No
field string The field related to the error ❌ No
path string The path of the field related to the error ❌ No