> ## Documentation Index
> Fetch the complete documentation index at: https://docs.base44.com/llms.txt
> Use this file to discover all available pages before exploring further.
# List agents
> This API is in beta. Endpoints, fields, and behavior may still change, so avoid depending on it in production.
Retrieves a list of AI agents in a [workspace](/documentation/account-and-billing/about-workspaces), including [Superagents](/Getting-Started/superagent).
Use this method to view, for each agent:
- Whether it's a [Superagent](/Getting-Started/superagent) or belongs to an [app](/Building-your-app/AI-agents-for-apps).
- Whether people can use it without signing in.
- Which AI model it uses.
- The channels people can use to reach it.
To call this method, your API key must include the `compliance:inventory:read` [scope](/developers/references/compliance-api/get-started/authentication#scopes).
## OpenAPI
````yaml /developers/references/compliance-api/compliance-openapi.json get /{workspace_id}/agents
openapi: 3.1.0
info:
title: Base44 Compliance API
description: >
Read-only API for security partners to review the security configuration of
a Base44 workspace.
The API reports what exists in a workspace and how it's configured. It
doesn't report activity. To review activity, use the Audit Logs API.
The Compliance API is available for Enterprise workspaces only. To get
started, see the
[overview](/developers/references/compliance-api/get-started/overview).
version: 1.0.0
servers:
- url: https://app.base44.com/api/v1/compliance
description: Compliance API
security:
- ApiKeyAuth: []
PartnerTokenAuth: []
tags:
- name: compliance
description: Read workspace configuration, inventory, and security posture.
paths:
/{workspace_id}/agents:
get:
tags:
- compliance
summary: List agents
description: >-
This API is in beta. Endpoints, fields, and behavior may still
change, so avoid depending on it in production.
Retrieves a list of AI agents in a
[workspace](/documentation/account-and-billing/about-workspaces),
including [Superagents](/Getting-Started/superagent).
Use this method to view, for each agent:
- Whether it's a [Superagent](/Getting-Started/superagent) or belongs to
an [app](/Building-your-app/AI-agents-for-apps).
- Whether people can use it without signing in.
- Which AI model it uses.
- The channels people can use to reach it.
To call this method, your API key must include the
`compliance:inventory:read`
[scope](/developers/references/compliance-api/get-started/authentication#scopes).
operationId: list_agents__workspace_id__agents_get
parameters:
- name: workspace_id
in: path
required: true
schema:
type: string
description: >-
Workspace ID. Learn more about how to [retrieve your workspace
ID](/developers/references/compliance-api/get-started/authentication#workspace-id).
title: Workspace Id
description: >-
Workspace ID. Learn more about how to [retrieve your workspace
ID](/developers/references/compliance-api/get-started/authentication#workspace-id).
example: 67f2c8e01a3b5d004e92d7a1
- name: cursor
in: query
required: false
schema:
anyOf:
- type: string
- type: 'null'
description: >-
Cursor for the next page. Pass the `next_cursor` value from the
previous response back unchanged. Don't send `limit` with it,
because the cursor already carries the limit. Cursors are valid
for 24 hours, and a cursor from a different workspace isn't valid.
Sending `limit` with a cursor, or an invalid cursor, returns a
`400` error.
title: Cursor
description: >-
Cursor for the next page. Pass the `next_cursor` value from the
previous response back unchanged. Don't send `limit` with it,
because the cursor already carries the limit. Cursors are valid for
24 hours, and a cursor from a different workspace isn't valid.
Sending `limit` with a cursor, or an invalid cursor, returns a `400`
error.
example: gAAAAABn7vJ...
- name: limit
in: query
required: false
schema:
anyOf:
- type: integer
maximum: 1000
minimum: 1
- type: 'null'
description: >-
Number of items per page. The maximum is 1000. Don't send it with
`cursor`, because the cursor already carries the limit.
Default: 100.
title: Limit
description: >-
Number of items per page. The maximum is 1000. Don't send it with
`cursor`, because the cursor already carries the limit.
Default: 100.
example: 100
responses:
'200':
description: Successful Response
content:
application/json:
schema:
$ref: '#/components/schemas/CompliancePage_ComplianceAgent_'
'400':
description: >-
The request is invalid, for example a parameter is malformed or the
`cursor` is invalid or expired. The error `type` is
`invalid_request_error`. See [error
codes](/developers/references/compliance-api/get-started/error-codes).
content:
application/json:
examples:
invalid_request_error:
summary: invalid_request_error
value:
error:
type: invalid_request_error
message: The request is invalid.
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
'401':
description: >-
A credential failed. The workspace API key is missing, unknown, or
disabled, or your signed token is missing or Base44 can't verify it
(`partner_key_required`). Check the `code` to see which credential
to fix. Don't ask the workspace admin to create a new key for
`partner_key_required`. The error `type` is `authentication_error`.
See [error
codes](/developers/references/compliance-api/get-started/error-codes).
content:
application/json:
examples:
authentication_error:
summary: authentication_error
value:
error:
type: authentication_error
message: Invalid workspace API key.
partner_key_required:
summary: partner_key_required
value:
error:
type: authentication_error
message: >-
Missing or invalid compliance partner assertion in the
Authorization header. The workspace API key is sent
separately and is not the credential at fault here.
code: partner_key_required
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
'402':
description: >-
The workspace's plan doesn't include the Compliance API, or its
subscription has lapsed (`plan_not_included`). The error `type` is
`permission_error`. See [error
codes](/developers/references/compliance-api/get-started/error-codes).
content:
application/json:
examples:
plan_not_included:
summary: plan_not_included
value:
error:
type: permission_error
message: This API key is not permitted to perform that request.
code: plan_not_included
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
'403':
description: >-
The key doesn't have the scope the method needs (`missing_scope`),
the key belongs to a different workspace (`workspace_mismatch`), the
Compliance API isn't enabled for the workspace
(`feature_not_enabled`), or Base44 suspended your partner
registration (`partner_not_approved`). The error `type` is
`permission_error`. See [error
codes](/developers/references/compliance-api/get-started/error-codes).
content:
application/json:
examples:
missing_scope:
summary: missing_scope
value:
error:
type: permission_error
message: >-
This API key is not permitted to perform that request.
Needed: [compliance:members:read]
code: missing_scope
required_scope: compliance:members:read
workspace_mismatch:
summary: workspace_mismatch
value:
error:
type: permission_error
message: API key does not belong to this workspace
code: workspace_mismatch
feature_not_enabled:
summary: feature_not_enabled
value:
error:
type: permission_error
message: The Compliance API is not enabled for this workspace.
code: feature_not_enabled
partner_not_approved:
summary: partner_not_approved
value:
error:
type: permission_error
message: >-
This compliance partner is not approved to call the
Compliance API. Contact Base44; the customer cannot
grant this.
code: partner_not_approved
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
'404':
description: >-
The workspace doesn't exist. The error `type` is `not_found_error`.
See [error
codes](/developers/references/compliance-api/get-started/error-codes).
content:
application/json:
examples:
not_found_error:
summary: not_found_error
value:
error:
type: not_found_error
message: Not found.
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
'429':
description: >-
The workspace sent too many requests. Wait the number of seconds in
the `retry-after` header, then retry with a newly signed
authentication token. The error `type` is `rate_limit_error`. See
[rate
limits](/developers/references/compliance-api/get-started/rate-limits).
content:
application/json:
examples:
rate_limit_error:
summary: rate_limit_error
value:
error:
type: rate_limit_error
message: >-
Rate limit exceeded. Retry after the interval in the
retry-after header.
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
'500':
description: >-
Something went wrong on Base44's side. Retry with exponential
backoff and a newly signed authentication token. If the code is
`inventory_too_large`, the workspace has more items than the API can
return, and retrying doesn't help. Ask your Base44 partner contact
for help. See [error
codes](/developers/references/compliance-api/get-started/error-codes).
content:
application/json:
examples:
api_error:
summary: api_error
value:
error:
type: api_error
message: >-
An internal error occurred. Retry with exponential
backoff.
inventory_too_large:
summary: inventory_too_large
value:
error:
type: api_error
message: >-
This workspace holds more rows than a single compliance
read supports (9999). Returning a partial inventory
would misreport it, so the request failed instead. Not
retryable without a change on our side.
code: inventory_too_large
schema:
$ref: '#/components/schemas/ComplianceErrorResponse'
components:
schemas:
CompliancePage_ComplianceAgent_:
properties:
data:
items:
$ref: '#/components/schemas/ComplianceAgent'
type: array
title: Data
description: Items on this page.
pagination:
$ref: '#/components/schemas/PublicPaginationInfo'
description: Pagination details for this page.
type: object
required:
- data
- pagination
title: CompliancePage[ComplianceAgent]
description: One page of results from a list endpoint.
ComplianceErrorResponse:
properties:
error:
$ref: '#/components/schemas/ComplianceErrorBody'
type: object
required:
- error
title: ComplianceErrorResponse
ComplianceAgent:
properties:
id:
type: string
title: Id
description: Unique ID of the agent, in the format `{app_id}:{agent_name}`.
example: 68a1f3c2b7d9e4005c20e8f6:support_agent
app_id:
type: string
title: App Id
description: >-
Unique ID of the app the agent belongs to. A Superagent is its own
app.
example: 68a1f3c2b7d9e4005c20e8f6
agent_name:
type: string
title: Agent Name
description: Name of the agent.
example: support_agent
standalone:
type: boolean
title: Standalone
description: >-
Whether the agent is a [Superagent](/Getting-Started/superagent). A
value of `false` indicates an agent that belongs to another app.
example: false
lifecycle:
type: string
title: Lifecycle
description: >-
Stage of the agent's setup. Possible values are `"active"` and
`"onboarding"`.
example: active
allow_anonymous_access:
anyOf:
- type: boolean
- type: 'null'
title: Allow Anonymous Access
description: >-
Whether people can use the agent without signing in. A `null` value
indicates the agent is a Superagent. Who can use a Superagent
depends on its app's visibility and sign-in settings. See the [List
apps](/api-reference/compliance/list-apps) method.
example: false
model:
anyOf:
- type: string
- type: 'null'
title: Model
description: >-
AI model the agent uses. A `null` value indicates the agent is a
Superagent, which doesn't use a fixed model.
example: claude-sonnet-5-5
channels:
items:
$ref: '#/components/schemas/ComplianceAgentChannel'
type: array
title: Channels
description: >-
Channels people can use to reach the agent. The list is the same for
every agent, because Base44 doesn't record which channels each agent
uses.
type: object
required:
- id
- app_id
- agent_name
- standalone
- lifecycle
- channels
title: ComplianceAgent
PublicPaginationInfo:
properties:
total:
type: integer
title: Total
description: Total number of matching items.
example: 347
next_cursor:
anyOf:
- type: string
- type: 'null'
title: Next Cursor
description: >-
Cursor for fetching the next page. The value is `null` if there are
no more pages.
example: gAAAAABn7vJ...
has_more:
type: boolean
title: Has More
description: Whether there are more items to fetch.
example: true
type: object
required:
- total
- has_more
title: PublicPaginationInfo
description: Pagination metadata for a cursor-paginated page.
ComplianceErrorBody:
properties:
type:
type: string
title: Type
description: >-
Category of the error. See [error
types](/developers/references/compliance-api/get-started/error-codes#error-types).
example: permission_error
message:
type: string
title: Message
description: Explanation of the error for people to read.
example: >-
This API key is not permitted to perform that request. Needed:
[compliance:members:read]
code:
anyOf:
- type: string
- type: 'null'
title: Code
description: >-
Specific cause of the error, such as `missing_scope`,
`workspace_mismatch`, `plan_not_included`, `feature_not_enabled`,
`partner_key_required`, `partner_not_approved`, or
`inventory_too_large`. Omitted when `type` is enough. See
[Compliance API
codes](/developers/references/compliance-api/get-started/error-codes#compliance-api-codes).
example: missing_scope
required_scope:
anyOf:
- type: string
- type: 'null'
title: Required Scope
description: >-
Scope the method needs. Included only when `code` is
`missing_scope`.
example: compliance:members:read
type: object
required:
- type
- message
title: ComplianceErrorBody
ComplianceAgentChannel:
properties:
type:
type: string
title: Type
description: >-
Type of channel. Possible values are `"whatsapp"`, `"telegram"`,
`"line"`, `"imessage"`, `"slack"`, `"partner_clients"`, and
`"phone"`.
example: slack
configured:
anyOf:
- type: boolean
- type: 'null'
title: Configured
description: >-
Whether the agent has the channel set up. A `null` value indicates
Base44 has no status for the channel. It doesn't indicate the
channel is off.
active:
anyOf:
- type: boolean
- type: 'null'
title: Active
description: >-
Whether people can currently reach the agent through the channel. A
`null` value indicates Base44 has no status for the channel. It
doesn't indicate the channel is off.
type: object
required:
- type
title: ComplianceAgentChannel
securitySchemes:
ApiKeyAuth:
type: apiKey
in: header
name: api_key
description: >-
Identifies the workspace you can read and limits your access to the
key's scopes. A workspace owner or admin creates the key and shares it
with you. See [Workspace API
key](/developers/references/compliance-api/get-started/authentication#workspace-api-key)
for more details.
PartnerTokenAuth:
type: http
scheme: bearer
bearerFormat: JWT
description: >-
The signed authentication token that identifies you as the security
partner. Sign a new token for each request with your signing key, using
`RS256` or `ES256`, and send it as a bearer token in the `Authorization`
header. See [Sign an authentication
token](/developers/references/compliance-api/get-started/authentication#sign-an-authentication-token)
for more details.
````
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.