GET /workflows/:workflowId

This page covers two workflow read endpoints.

Purpose	Endpoint
Full workflow definition	`GET /v1/workflows/{workflowId}`
Lightweight info for lists	`GET /v1/workflows/{workflowId}/summary`

Creating and editing workflows is not available through the API. Use the dashboard.

Get workflow detail

GET /v1/workflows/{workflowId}

curl "https://client-omni-api.argosidentity.com/v1/workflows/wf_3p8k2m5r9v1t" \
  -H "x-api-key: your-api-key-here"

Response fields

Field	Type	Description
`id`	`string`	Workflow ID (`wf_` prefix)
`projectId`	`string`	Parent project ID (`proj_` prefix)
`name`	`string`	Workflow name
`description`	`string \| null`	Description
`status`	`string`	`active` · `archived`
`policyText`	`string \| null`	Full policy text you entered. The playbook is generated from this text
`policyDocuments`	`object[] \| null`	Metadata for attached policy documents. Empty array when there are none
`playbookId`	`string \| null`	Playbook ID (`PB-` prefix). Stays the same when you change the policy and regenerate
`playbookVersion`	`string \| null`	Playbook version. Increments on every edit
`playbookContent`	`string \| null`	Playbook body (Markdown)
`workflowActions`	`object[]`	Every action the playbook defines. See below
`referenceNotes`	`string[]`	A read-only derived value — the `referenceNotes` of every action, deduplicated and merged
`engines`	`object[]`	The system-wide engine catalog. See below
`outputSchema`	`object \| null`	JSON Schema (draft-07) defining the analysis result structure
`aiModelId`	`string \| null`	ID of the selected AI model. `null` when none is selected, and the default model runs
`aiModelProvider`	`string \| null`	Provider identifier of the selected model. `null` means the default model
`aiModel`	`object \| null`	Catalog info for the selected model. See below
`profilesCount`	`number`	Number of linked profiles
`totalAnalyses`	`number`	Cumulative analysis count
`createdAt`	`string`	Creation time
`updatedAt`	`string`	Last update time

`workflowActions` action definitions

Every action the playbook defines. agentAuditLog.steps in an analysis response holds only the actions that actually ran, so use this array as the denominator when you compute an execution rate.

Field	Type	Description
`step`	`number`	Execution order (starts at 1). Reordering renumbers from the top, so do not use it as an identifier
`workId`	`number`	Stable identifier for the action. It survives reordering and matches `workId` and `agentAuditLog.steps[].work_id` in the analysis response
`actionName`	`string`	Action function name (`^[a-z][a-z0-9_]*$`)
`description`	`string`	Action description
`referenceNotes`	`string[]`	Reference notes for this action, carrying the grounds from the original policy. Empty array when there are none
`engines`	`object[]`	Engines assigned to this action. Empty array when none are assigned

Action names are generated from your policy text and are not deterministic — regenerating a playbook from the same policy can produce different names. Match actions by workId, never by actionName.

`workflowActions.engines`

Field	Type	Description
`engineId`	`string`	Engine ID. Matches `id` in the `engines` catalog
`type`	`string`	Engine type
`name`	`string`	Engine display name

`engines` engine catalog

The full list of engines this workflow can choose from. isSelected tells you which ones it actually uses.

Field	Type	Description
`id`	`string`	Engine ID
`code`	`string`	Engine code
`name`	`string`	Engine display name
`description`	`string \| null`	Engine description
`executionOrder`	`number`	Display order in the catalog
`isSelected`	`boolean`	Whether an action in this workflow actually uses it
`isRequired`	`boolean`	`true` when any action marks it as required
`pricePerCall`	`number`	Credit price per call

`policyDocuments`

Field	Type	Description
`name`	`string`	Uploaded file name
`mimeType`	`string`	MIME type
`size`	`number`	File size in bytes
`uploadedAt`	`string`	Upload time
`s3Key`	`string`	Internal storage reference key. Do not use it in your integration logic

`aiModel`

Field	Type	Description
`modelId`	`string`	Model identifier
`displayName`	`string`	Model display name
`provider`	`string`	Provider identifier
`hostingLabel`	`string`	Hosting label
`description`	`string`	Model description
`isDefault`	`boolean`	Whether this is the default model

For the models you can choose, see AI model selection.

Example response

policyText, playbookContent, and outputSchema are much longer in practice, and workflowActions and engines hold more entries than shown here.

{
  "id": "wf_3p8k2m5r9v1t",
  "projectId": "proj_75hgxoxamzn9",
  "name": "Vendor Onboarding Document Verification Policy",
  "description": "This is an example for KYB",
  "status": "active",
  "policyText": "Vendor Onboarding Document Verification Policy\n\n1. Required documents\n- A certificate of good standing, a bank verification letter, and a beneficial ownership certification must all be submitted ...",
  "policyDocuments": [],
  "playbookId": "PB-20260910-VM351L",
  "playbookVersion": "1.0.0",
  "playbookContent": "---\nname: playbook.md\ndescription: Vendor onboarding document verification\n---\n...",
  "workflowActions": [
    {
      "step": 1,
      "workId": 1,
      "actionName": "verify_document_presence",
      "description": "Check that all three required documents were submitted",
      "referenceNotes": [
        "All three documents must be submitted (policy 1)",
        "If any one is missing, mark the case for review (policy 1)"
      ],
      "engines": [
        { "engineId": "eng_1a2b3c4d5e6f", "type": "text-similarity", "name": "Text Verifier" }
      ]
    },
    {
      "step": 6,
      "workId": 6,
      "actionName": "record_and_validate_owners",
      "description": "Record beneficial owners and check that ownership percentages do not exceed 100%",
      "referenceNotes": ["If ownership percentages total more than 100%, mark the case for review (policy 6)"],
      "engines": []
    }
  ],
  "referenceNotes": [
    "All three documents must be submitted (policy 1)",
    "If ownership percentages total more than 100%, mark the case for review (policy 6)"
  ],
  "engines": [
    {
      "id": "eng_1a2b3c4d5e6f",
      "code": "text-similarity",
      "name": "Text Verifier",
      "description": "Compares text across documents to check that it matches",
      "executionOrder": 1,
      "isSelected": true,
      "isRequired": false,
      "pricePerCall": 0.02
    },
    {
      "id": "eng_3m4n5o6p7q8r",
      "code": "aml-business",
      "name": "AML - Business",
      "description": "Screens a legal entity name against AML and sanctions lists",
      "executionOrder": 7,
      "isSelected": true,
      "isRequired": false,
      "pricePerCall": 0.15
    }
  ],
  "outputSchema": {
    "type": "object",
    "properties": {
      "company": {
        "type": "object",
        "description": "Company details extracted from the certificate of good standing",
        "properties": {
          "legal_name": { "type": "string", "description": "Registered legal name" },
          "state_file_number": { "type": "string", "description": "State file number" }
        }
      },
      "documents_complete": { "type": "boolean", "description": "Whether all three required documents were submitted" }
    }
  },
  "aiModelId": "<selected model ID>",
  "aiModelProvider": "<provider identifier>",
  "aiModel": {
    "modelId": "<selected model ID>",
    "displayName": "<model display name>",
    "provider": "<provider identifier>",
    "hostingLabel": "<hosting label>",
    "description": "<model description>",
    "isDefault": false
  },
  "profilesCount": 1,
  "totalAnalyses": 1,
  "createdAt": "2026-09-10T05:10:22.000Z",
  "updatedAt": "2026-09-10T05:14:07.000Z"
}

The values inside aiModel depend on the model you pick in the dashboard, so they are placeholders above. When you do not select an AI model, aiModelId and aiModelProvider are null and the default model runs.

Get workflow summary

GET /v1/workflows/{workflowId}/summary

Returns metadata only, without the policy, playbook, or schema bodies. Use it for lightweight reads such as list screens.

curl "https://client-omni-api.argosidentity.com/v1/workflows/wf_3p8k2m5r9v1t/summary" \
  -H "x-api-key: your-api-key-here"

Field	Type	Description
`id`	`string`	Workflow ID
`name`	`string`	Workflow name
`description`	`string \| null`	Description
`playbookId`	`string \| null`	Playbook ID
`profilesCount`	`number`	Number of linked profiles
`totalAnalyses`	`number`	Cumulative analysis count
`status`	`string`	Workflow status

{
  "id": "wf_3p8k2m5r9v1t",
  "name": "Vendor Onboarding Document Verification Policy",
  "description": "This is an example for KYB",
  "playbookId": "PB-20260910-VM351L",
  "profilesCount": 1,
  "totalAnalyses": 1,
  "status": "active"
}

Error codes

Status	Code	Description
404	`OMNI_3003`	Workflow not found

Common errors (400 validation · 401 authentication · 403 permission · 500 server) are listed in the API overview.

Overview

API Reference

GET /workflows/:workflowId

Get workflow detail

Response fields

`workflowActions` action definitions

`workflowActions.engines`

`engines` engine catalog

`policyDocuments`

`aiModel`

Example response

Get workflow summary

Error codes

​Get workflow detail

​Response fields

​workflowActions action definitions

​workflowActions.engines

​engines engine catalog

​policyDocuments

​aiModel

​Example response

​Get workflow summary

​Error codes

Get workflow detail

Response fields

`workflowActions` action definitions

`workflowActions.engines`

`engines` engine catalog

`policyDocuments`

`aiModel`

Example response

Get workflow summary

Error codes