> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to Knock

The Knock connector lets your AI agent act in each user's Knock account. Each user connects their own Knock access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 54 tools.

**Authentication:** Bearer Token
**Categories:** Communication, Developer Tools
**Tools:** 54: 26 read, 19 write, 9 destructive
**Users sign in with:** Access token
**Built by:** Scalekit connector
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=KNOCK)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the Knock connection

   In **AgentKit > Connections**, create a Knock connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'knock'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Knock:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'knock_messages_list',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "knock"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize Knock:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="knock_messages_list",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their Knock credentials there. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'knock_audience_members_list',
  toolInput: {
    key: '<key>',
  },
  connector: 'knock',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="knock_audience_members_list",
    tool_input={
        "key": "<key>",
    },
    connection_name="knock",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `knock_audience_members_list`

List Audience Members · Read-only

Return a paginated list of members for the specified audience.
Returns entries, an array of items with __typename, added_at, tenant, user, user_id, plus page_info.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `key` (`string`, required): The key of the audience. Example: new-comment

Example input: `{"key":"<key>"}`

### `knock_bulk_operation_get`

Get bulk operation · Read-only

Retrieve a bulk operation (if it exists) and displays the current state of it.
Returns an object with __typename, completed_at, error_count, error_items, estimated_total_rows, failed_at, id, inserted_at, name, processed_rows, progress_path, started_at and more.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): The ID of the bulk operation to retrieve. Example: 3b6b2e4c-1d2a-4f5e-8a9b-0c1d2e3f4a5b

Example input: `{"id":"<id>"}`

### `knock_message_activities_list`

List Message Activities · Read-only

Return a paginated list of activities for the specified message.
Returns items, an array of items with __typename, actor, data, id, inserted_at, recipient, updated_at, plus page_info.
Needs message_id from knock_messages_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The ID of the message to fetch activities for. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `trigger_data` (`string`, optional): The trigger data to filter activities by. Example: {"order_id": "A-1001"}

Example input: `{"message_id":"<message_id>"}`

### `knock_message_content_get`

Get message content · Read-only

Return the fully rendered contents of a message, where the response depends on which channel the message was sent through.
Returns an object with __typename, data, inserted_at, message_id.
Needs message_id from knock_messages_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The ID of the message to fetch contents of. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7

Example input: `{"message_id":"<message_id>"}`

### `knock_message_delivery_logs_list`

List Message Delivery Logs · Read-only

Return a paginated list of delivery logs for the specified message.
Returns items, an array of items with __typename, environment_id, id, inserted_at, request, response, service_name, plus page_info.
Needs message_id from knock_messages_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The ID of the message to fetch delivery logs for. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10

Example input: `{"message_id":"<message_id>"}`

### `knock_message_events_list`

List Message Events · Read-only

Return a paginated list of events for the specified message.
Returns items, an array of items with __typename, data, id, inserted_at, recipient, type, plus page_info.
Needs message_id from knock_messages_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The ID of the message to fetch events for. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10

Example input: `{"message_id":"<message_id>"}`

### `knock_message_get`

Get message · Read-only

Retrieve a specific message by its ID.
Returns an object with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, interacted_at, link_clicked_at and more.
Use this when you have the ID. Use knock_messages_list to find one.
Needs message_id from knock_messages_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The unique identifier for the message. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7

Example input: `{"message_id":"<message_id>"}`

### `knock_messages_list`

List messages · Read-only

Return a paginated list of messages for the current environment.
Returns items, an array of items with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, plus page_info.
Use this to browse. Use knock_message_get when you already have the ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `channel_id` (`string`, optional): Limits the results to items with the corresponding channel ID. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `engagement_status` (`array`, optional): Limits the results to messages with the given engagement status. Example: ["unread"]
- `inserted_at_gt` (`string`, optional): Limits the results to items inserted after the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_gte` (`string`, optional): Limits the results to items inserted after or on the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_lt` (`string`, optional): Limits the results to items inserted before the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_lte` (`string`, optional): Limits the results to items inserted before or on the given date. Example: 2025-01-01T00:00:00Z
- `message_ids` (`array`, optional): Limits the results to only the message IDs given (max 50). Note: when using this option, the results will be subject to any other filters applied to the query. Example: ["1jNaXzB2RZX3LY8wVQnfCKyPnv7"]
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `source` (`string`, optional): Limits the results to messages triggered by the given workflow key. Example: comment-created
- `status` (`array`, optional): Limits the results to messages with the given delivery status. Example: ["delivered"]
- `tenant` (`string`, optional): Limits the results to items with the corresponding tenant. Example: acme-corp
- `trigger_data` (`string`, optional): Limits the results to only messages that were generated with the given data. See trigger data filtering for more information. Example: {"order_id": "A-1001"}
- `workflow_categories` (`array`, optional): Limits the results to messages related to any of the provided categories. Example: ["workflow_123"]
- `workflow_recipient_run_id` (`string`, optional): Limits the results to messages for a specific recipient's workflow run. Example: 123e4567-e89b-12d3-a456-426614174000
- `workflow_run_id` (`string`, optional): Limits the results to messages associated with the top-level workflow run ID returned by the workflow trigger request. Example: 123e4567-e89b-12d3-a456-426614174000

Example input: `{}`

### `knock_object_channel_data_get`

Get Object Channel Data · Read-only

Return the channel data for the specified object and channel.
Returns an object with __typename, channel_id, data, provider.
Needs object_id from knock_objects_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `channel_id` (`string`, required): The unique identifier for the channel. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `object_id` (`string`, required): Unique identifier for the object. Example: project-123

Example input: `{"channel_id":"<channel_id>","collection":"<collection>","object_id":"<object_id>"}`

### `knock_object_get`

Get an object · Read-only

Retrieve a specific object by its ID from the specified collection.
Returns an object with __typename, collection, created_at, id, properties, updated_at.
Use this when you have the ID. Use knock_objects_list to find one.
Needs id from knock_objects_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `id` (`string`, required): Unique identifier for the object. Example: project-123

Example input: `{"collection":"<collection>","id":"<id>"}`

### `knock_object_messages_list`

List Object Messages · Read-only

Return a paginated list of messages for a specific object in the given collection.
Returns items, an array of items with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, plus page_info.
Needs id from knock_objects_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `id` (`string`, required): Unique identifier for the object. Example: project-123
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `channel_id` (`string`, optional): Limits the results to items with the corresponding channel ID. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `engagement_status` (`array`, optional): Limits the results to messages with the given engagement status. Example: ["unread"]
- `inserted_at_gt` (`string`, optional): Limits the results to items inserted after the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_gte` (`string`, optional): Limits the results to items inserted after or on the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_lt` (`string`, optional): Limits the results to items inserted before the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_lte` (`string`, optional): Limits the results to items inserted before or on the given date. Example: 2025-01-01T00:00:00Z
- `message_ids` (`array`, optional): Limits the results to only the message IDs given (max 50). Note: when using this option, the results will be subject to any other filters applied to the query. Example: ["1jNaXzB2RZX3LY8wVQnfCKyPnv7"]
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `source` (`string`, optional): Limits the results to messages triggered by the given workflow key. Example: comment-created
- `status` (`array`, optional): Limits the results to messages with the given delivery status. Example: ["delivered"]
- `tenant` (`string`, optional): Limits the results to items with the corresponding tenant. Example: acme-corp
- `trigger_data` (`string`, optional): Limits the results to only messages that were generated with the given data. See trigger data filtering for more information. Example: {"order_id": "A-1001"}
- `workflow_categories` (`array`, optional): Limits the results to messages related to any of the provided categories. Example: ["workflow_123"]
- `workflow_recipient_run_id` (`string`, optional): Limits the results to messages for a specific recipient's workflow run. Example: 123e4567-e89b-12d3-a456-426614174000
- `workflow_run_id` (`string`, optional): Limits the results to messages associated with the top-level workflow run ID returned by the workflow trigger request. Example: 123e4567-e89b-12d3-a456-426614174000

Example input: `{"collection":"<collection>","id":"<id>"}`

### `knock_object_subscriptions_list`

List subscriptions · Read-only

List subscriptions for an object. Either list the recipients that subscribe to the provided object, or list the objects that the provided object is subscribed to.
Returns entries, an array of items with __typename, inserted_at, object, properties, recipient, updated_at, plus page_info.
Needs object_id from knock_objects_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `object_id` (`string`, required): Unique identifier for the object. Example: project-123
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `include` (`array`, optional): Additional fields to include in the response. Example: ["preferences"]
- `mode` (`string`, optional): Mode of the request. `recipient` to list the objects that the provided object is subscribed to, `object` to list the recipients that subscribe to the provided object. Example: object. One of: `recipient`, `object`.
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10

Example input: `{"collection":"<collection>","object_id":"<object_id>"}`

### `knock_objects_list`

List objects in a collection · Read-only

Return a paginated list of objects from the specified collection.
Returns entries, an array of items with __typename, collection, created_at, id, properties, updated_at, plus page_info.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `include` (`array`, optional): Includes preferences of the objects in the response. Example: ["preferences"]
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10

Example input: `{"collection":"<collection>"}`

### `knock_schedules_list`

List schedules · Read-only

Return a paginated list of schedules for the current environment, filtered by workflow and optionally by recipients and tenant.
Returns entries, an array of items with __typename, actor, data, id, inserted_at, last_occurrence_at, next_occurrence_at, recipient, repeats, tenant, plus page_info.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `workflow` (`string`, required): Filter by workflow key. Example: new-comment
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `recipients__` (`array`, optional): Filter by recipient references. Example: ["user_123"]
- `tenant` (`string`, optional): Filter by tenant ID. Example: acme-corp

Example input: `{"workflow":"<workflow>"}`

### `knock_tenant_get`

Get a tenant · Read-only

Get a tenant by ID.
Returns an object with __typename, id, name, settings.
Use this when you have the ID. Use knock_tenants_list to find one.
Needs id from knock_tenants_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): The unique identifier for the tenant. Example: acme-corp
- `resolve_full_preference_settings` (`boolean`, optional): When true, merges environment-level default preferences into the tenant's `settings.preference_set` field before returning the response. Defaults to false. Example: false

Example input: `{"id":"<id>"}`

### `knock_tenants_list`

List tenants · Read-only

List tenants for the current environment.
Returns entries, an array of items with __typename, id, name, settings, plus page_info.
Use this to browse. Use knock_tenant_get when you already have the ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `name` (`string`, optional): Filter tenants by name. Example: Scalekit test
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `tenant_id` (`string`, optional): Filter tenants by ID. Example: user-123

Example input: `{}`

### `knock_user_channel_data_get`

Get User Channel Data · Read-only

Retrieve the channel data for a specific user and channel ID.
Returns an object with __typename, channel_id, data, provider.
Needs user_id from knock_users_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `channel_id` (`string`, required): The unique identifier for the channel. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `user_id` (`string`, required): The unique identifier of the user. Example: user-123

Example input: `{"channel_id":"<channel_id>","user_id":"<user_id>"}`

### `knock_user_get`

Get user · Read-only

Retrieve a specific user by their ID.
Returns an object with __typename, avatar, created_at, email, id, name, phone_number, timezone, updated_at.
Use this when you have the ID. Use knock_users_list to find one.
Needs user_id from knock_users_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The ID of the user to retrieve. Example: user-123

Example input: `{"user_id":"<user_id>"}`

### `knock_user_messages_list`

List User Messages · Read-only

Return a paginated list of messages for a specific user.
Returns items, an array of items with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, plus page_info.
Needs user_id from knock_users_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The user ID to list messages for. Example: user-123
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `channel_id` (`string`, optional): Limits the results to items with the corresponding channel ID. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `engagement_status` (`array`, optional): Limits the results to messages with the given engagement status. Example: ["unread"]
- `inserted_at_gt` (`string`, optional): Limits the results to items inserted after the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_gte` (`string`, optional): Limits the results to items inserted after or on the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_lt` (`string`, optional): Limits the results to items inserted before the given date. Example: 2025-01-01T00:00:00Z
- `inserted_at_lte` (`string`, optional): Limits the results to items inserted before or on the given date. Example: 2025-01-01T00:00:00Z
- `message_ids` (`array`, optional): Limits the results to only the message IDs given (max 50). Note: when using this option, the results will be subject to any other filters applied to the query. Example: ["1jNaXzB2RZX3LY8wVQnfCKyPnv7"]
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `source` (`string`, optional): Limits the results to messages triggered by the given workflow key. Example: comment-created
- `status` (`array`, optional): Limits the results to messages with the given delivery status. Example: ["delivered"]
- `tenant` (`string`, optional): Limits the results to items with the corresponding tenant. Example: acme-corp
- `trigger_data` (`string`, optional): Limits the results to only messages that were generated with the given data. See trigger data filtering for more information. Example: {"order_id": "A-1001"}
- `workflow_categories` (`array`, optional): Limits the results to messages related to any of the provided categories. Example: ["workflow_123"]
- `workflow_recipient_run_id` (`string`, optional): Limits the results to messages for a specific recipient's workflow run. Example: 123e4567-e89b-12d3-a456-426614174000
- `workflow_run_id` (`string`, optional): Limits the results to messages associated with the top-level workflow run ID returned by the workflow trigger request. Example: 123e4567-e89b-12d3-a456-426614174000

Example input: `{"user_id":"<user_id>"}`

### `knock_user_preference_set_get`

Get user preference set · Read-only

Retrieve a specific preference set for a user identified by the preference set ID.
Returns an object with categories, channel_types, channels, commercial_subscribed, id, workflows.
Use this when you have the ID. Use knock_user_preference_sets_list to find one.
Needs user_id from knock_users_list; id from knock_user_preference_sets_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): Unique identifier for the preference set. Example: default
- `user_id` (`string`, required): The unique identifier of the user. Example: user-123
- `tenant` (`string`, optional): The unique identifier for the tenant. Example: acme-corp

Example input: `{"id":"<id>","user_id":"<user_id>"}`

### `knock_user_preference_sets_list`

List user preference sets · Read-only

Retrieve a list of all preference sets for a specific user.
Returns an array of items with categories, channel_types, channels, commercial_subscribed, id, workflows.
Use this to browse. Use knock_user_preference_set_get when you already have the ID.
Needs user_id from knock_users_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The unique identifier of the user. Example: user-123

Example input: `{"user_id":"<user_id>"}`

### `knock_user_schedules_list`

List user schedules · Read-only

Return a paginated list of schedules for a specific user, in descending order.
Returns entries, an array of items with __typename, actor, data, id, inserted_at, last_occurrence_at, next_occurrence_at, recipient, repeats, tenant, plus page_info.
Needs user_id from knock_users_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The user ID to list schedules for. Example: user-123
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `tenant` (`string`, optional): The tenant ID to filter schedules for. Example: acme-corp
- `workflow` (`string`, optional): The workflow key to filter schedules for. Example: new-comment

Example input: `{"user_id":"<user_id>"}`

### `knock_user_subscriptions_list`

List user subscriptions · Read-only

Retrieve a paginated list of subscriptions for a specific user, in descending order.
Returns entries, an array of items with __typename, inserted_at, object, properties, recipient, updated_at, plus page_info.
Needs user_id from knock_users_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The user ID to list subscriptions for. Example: user-123
- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `include` (`array`, optional): Associated resources to include in the response. Example: ["preferences"]
- `objects__` (`array`, optional): Only returns subscriptions for the specified object references. Example: ["user_123"]
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10

Example input: `{"user_id":"<user_id>"}`

### `knock_users_list`

List users · Read-only

Retrieve a paginated list of users in the environment.
Returns entries, an array of items with __typename, avatar, created_at, email, id, name, phone_number, timezone, updated_at, plus page_info.
Use this to browse. Use knock_user_get when you already have the ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `include` (`array`, optional): Associated resources to include in the response. Example: ["preferences"]
- `order` (`string`, optional): Sort by insertion order. Pass `asc` for oldest first, or `desc` for newest first. Defaults to oldest first when omitted. Example: asc. One of: `asc`, `desc`.
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10

Example input: `{}`

### `knock_workflow_run_get`

Get Workflow Recipient Run · Read-only

Return a single workflow recipient run with its associated events.
Returns an object with __typename, actor, error_count, id, inserted_at, recipient, status, tenant, trigger_source, updated_at, workflow, workflow_run_id and more.
Use this when you have the ID. Use knock_workflow_runs_list to find one.
Needs id from knock_workflow_runs_list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): The unique identifier for the workflow recipient run (per-recipient). Example: 2Gm6S9mNOg9YIrd7NhbTmIqE7Qw

Example input: `{"id":"<id>"}`

### `knock_workflow_runs_list`

List Workflow Recipient Runs · Read-only

Return a paginated list of workflow recipient runs for the current environment.
Returns items, an array of items with __typename, actor, error_count, id, inserted_at, recipient, status, tenant, trigger_source, updated_at, plus page_info.
Use this to browse. Use knock_workflow_run_get when you already have the ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `after` (`string`, optional): The cursor to fetch entries after. Example: eyJpZCI6IjEyMyJ9
- `before` (`string`, optional): The cursor to fetch entries before. Example: eyJpZCI6IjEyMyJ9
- `ending_at` (`string`, optional): Limits the results to workflow recipient runs started before the given date. Example: 2025-01-01T00:00:00Z
- `has_errors` (`boolean`, optional): Limits the results to workflow recipient runs that have errors. Example: false
- `page_size` (`integer`, optional): The number of items per page (defaults to 50). Example: 10
- `recipient` (`object`, optional): Limits the results to workflow recipient runs for the given recipient. Accepts a user ID string or an object reference with `id` and `collection`. Example: user_123
- `starting_at` (`string`, optional): Limits the results to workflow recipient runs started after the given date. Example: 2025-01-01T00:00:00Z
- `status` (`array`, optional): Limits the results to workflow recipient runs with the given status. Example: ["queued"]
- `tenant` (`string`, optional): Limits the results to workflow recipient runs for the given tenant. Example: acme-corp
- `workflow` (`string`, optional): Limits the results to workflow recipient runs for the given workflow key. Example: new-comment

Example input: `{}`

### `knock_audience_members_add`

Add Audience Members · Write

Add one or more members to the specified audience.
Returns an empty response on success (HTTP 204).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `key` (`string`, required): The key of the audience. Example: new-comment
- `members` (`array`, required): A list of audience members to add. You can add up to 1,000 members per request. Example: [{"tenant": "ingen_isla_nublar", "user": {"email": "ellie@ingen.net", "id": "dr_sattler", "name": "Dr. Ellie Sattler"}}]
- `create_audience` (`boolean`, optional): Create the audience if it does not exist. Example: false

Example input: `{"key":"<key>","members":[{"tenant":"ingen_isla_nublar","user":{"email":"ellie@ingen.net","id":"dr_sattler","name":"Dr. Ellie Sattler"}}]}`

### `knock_message_archive`

Archive message · Write

Archive a message for the user. Archived messages are hidden from the default message list in the feed but can still be accessed and unarchived later.
Returns an object with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, interacted_at, link_clicked_at and more.
The request replaces the stored value with what you send.
Needs message_id from knock_messages_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The unique identifier for the message. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7

Example input: `{"message_id":"<message_id>"}`

### `knock_message_mark_interacted`

Mark message as interacted · Write

Mark a message as `interacted` with by the user. This can include any user action on the message, with optional metadata about the specific interaction.
Returns an object with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, interacted_at, link_clicked_at and more.
The request replaces the stored value with what you send.
Needs message_id from knock_messages_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The unique identifier for the message. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7
- `metadata` (`object`, optional): Metadata about the interaction. Example: {"action": "clicked_cta"}

Example input: `{"message_id":"<message_id>"}`

### `knock_message_mark_read`

Mark message as read · Write

Mark a message as `read`. This indicates that the user has read the message content.
Returns an object with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, interacted_at, link_clicked_at and more.
The request replaces the stored value with what you send.
Needs message_id from knock_messages_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The unique identifier for the message. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7

Example input: `{"message_id":"<message_id>"}`

### `knock_message_mark_seen`

Mark message as seen · Write

Mark a message as `seen`. This indicates that the user has viewed the message in their feed or inbox.
Returns an object with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, interacted_at, link_clicked_at and more.
The request replaces the stored value with what you send.
Needs message_id from knock_messages_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The unique identifier for the message. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7

Example input: `{"message_id":"<message_id>"}`

### `knock_message_unarchive`

Unarchive Message · Write

Remove a message from the archived state, making it visible in the default message list in the feed again.
Returns an object with __typename, actors, archived_at, channel, channel_id, clicked_at, data, engagement_statuses, id, inserted_at, interacted_at, link_clicked_at and more.
Needs message_id from knock_messages_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `message_id` (`string`, required): The unique identifier for the message. Example: 1jNaXzB2RZX3LY8wVQnfCKyPnv7

Example input: `{"message_id":"<message_id>"}`

### `knock_object_channel_data_set`

Set Object Channel Data · Write

Set the channel data for the specified object and channel.
Returns an object with __typename, channel_id, data, provider.
Needs object_id from knock_objects_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `channel_id` (`string`, required): The unique identifier for the channel. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `data` (`object`, required): Channel data for a given channel type. Example: {"tokens": ["fcm-token-abc123"]}
- `object_id` (`string`, required): Unique identifier for the object. Example: project-123

Example input: `{"channel_id":"<channel_id>","collection":"<collection>","data":{"tokens":["fcm-token-abc123"]},"object_id":"<object_id>"}`

### `knock_object_set`

Set Object · Write

Create an object or update an existing one in a collection (an upsert).
Returns an object with __typename, collection, created_at, id, properties, updated_at.
Needs id from knock_objects_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `id` (`string`, required): Unique identifier for the object. Example: project-123
- `channel_data` (`object`, optional): A request to set channel data for a type of channel inline. Example: {"97c5837d-c65c-4d54-aa39-080eeb81c69d": {"tokens": ["push_token_123"]}}
- `locale` (`string`, optional): The locale of the object. Used for message localization. Example: en-US
- `name` (`string`, optional): An optional name for the object. Example: My product
- `preferences` (`object`, optional): Inline set preferences for a recipient, where the key is the preference set id. Preferences that are set inline will be merged into any existing preferences rather than replacing them. Example: {"default": {"channel_types": {"email": true}, "workflows": {"dinosaurs-loose": {"channel_types": {"email": true}}}}}
- `timezone` (`string`, optional): The timezone of the object. Must be a valid tz database time zone string. Used for recurring schedules. Example: America/New_York

Example input: `{"collection":"<collection>","id":"<id>"}`

### `knock_object_subscriptions_add`

Add subscriptions · Write

Add subscriptions for an object. If a subscription already exists, it will be updated.
Returns an array of items with __typename, inserted_at, object, properties, recipient, updated_at.
Needs object_id from knock_objects_list.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `object_id` (`string`, required): Unique identifier for the object. Example: project-123
- `recipients` (`array`, required): The recipients of the subscription. You can subscribe up to 100 recipients to an object at a time. Example: ["user_1", "user_2"]
- `properties` (`object`, optional): The custom properties associated with the subscription relationship. Example: {"role": "owner"}

Example input: `{"collection":"<collection>","object_id":"<object_id>","recipients":["user_1","user_2"]}`

### `knock_objects_bulk_set`

Bulk Set Objects · Write

Bulk sets up to 1,000 objects at a time in the specified collection.
Returns an object with __typename, completed_at, error_count, error_items, estimated_total_rows, failed_at, id, inserted_at, name, processed_rows, progress_path, started_at and more.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `objects` (`array`, required): A list of objects. Example: [{"id": "project_1", "name": "My project"}]

Example input: `{"collection":"<collection>","objects":[{"id":"project_1","name":"My project"}]}`

### `knock_schedules_create`

Create Schedules · Write

Create one or more schedules for a workflow with the specified recipients, timing, and data.
Returns an array of items with __typename, actor, data, id, inserted_at, last_occurrence_at, next_occurrence_at, recipient, repeats, tenant.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `recipients` (`array`, required): The recipients to set the schedule for. Limited to 100 recipients per request. Example: ["user_123"]
- `workflow` (`string`, required): The key of the workflow. Example: new-comment
- `actor` (`object`, optional): A map of properties describing a user or an object to identify in Knock and mark as who or what performed the action. Example: {"id": "user_1"}
- `data` (`object`, optional): An optional map of data to pass into the workflow execution. There is a 10MB limit on the size of the full `data` payload. Any individual string value greater than 1024 bytes in length will be truncated in your logs. Example: {"order_id": "A-1001", "comment": "Looks good"}
- `ending_at` (`string`, optional): The ending date and time for the schedule. Example: 2026-01-15T09:30:00Z
- `repeats` (`array`, optional): The repeat rule for the schedule. Example: [{"__typename": "ScheduleRepeat", "day_of_month": null, "days": ["mon", "tue", "wed", "thu", "fri", "sat", "sun"], "frequency": "daily", "hours": null, "interval": 1, "minutes": null}]
- `scheduled_at` (`string`, optional): The starting date and time for the schedule. Example: 2026-01-15T09:30:00Z
- `tenant` (`object`, optional): The tenant to trigger the workflow for. Triggering with a tenant will use any tenant-level overrides associated with the tenant object, and all messages produced from workflow runs will be tagged with the tenant. Example: acme-corp

Example input: `{"recipients":["user_123"],"workflow":"<workflow>"}`

### `knock_schedules_update`

Update Schedules · Write

Update one or more existing schedules with new timing, data, or other properties.
Returns an array of items with __typename, actor, data, id, inserted_at, last_occurrence_at, next_occurrence_at, recipient, repeats, tenant.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `schedule_ids` (`array`, required): A list of schedule IDs. Example: ["123e4567-e89b-12d3-a456-426614174000"]
- `actor` (`object`, optional): A map of properties describing a user or an object to identify in Knock and mark as who or what performed the action. Example: user_123
- `data` (`object`, optional): An optional map of data to pass into the workflow execution. There is a 10MB limit on the size of the full `data` payload. Any individual string value greater than 1024 bytes in length will be truncated in your logs. Example: {"order_id": "A-1001", "comment": "Looks good"}
- `ending_at` (`string`, optional): The ending date and time for the schedule. Example: 2026-01-15T09:30:00Z
- `repeats` (`array`, optional): The repeat rule for the schedule. Example: [{"__typename": "ScheduleRepeat", "day_of_month": null, "days": ["mon", "tue", "wed", "thu", "fri", "sat", "sun"], "frequency": "daily", "hours": null, "interval": 1, "minutes": null}]
- `scheduled_at` (`string`, optional): The starting date and time for the schedule. Example: 2026-01-15T09:30:00Z
- `tenant` (`object`, optional): The tenant to trigger the workflow for. Triggering with a tenant will use any tenant-level overrides associated with the tenant object, and all messages produced from workflow runs will be tagged with the tenant. Example: acme-corp

Example input: `{"schedule_ids":["123e4567-e89b-12d3-a456-426614174000"]}`

### `knock_tenant_set`

Set a tenant · Write

Set a tenant within an environment, performing an upsert operation.
Returns an object with __typename, id, name, settings.
Needs id from knock_tenants_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): The unique identifier for the tenant. Example: acme-corp
- `channel_data` (`object`, optional): The channel data for the tenant. Example: {"97c5837d-c65c-4d54-aa39-080eeb81c69d": {"tokens": ["push_token_xxx"]}}
- `name` (`string`, optional): An optional name for the tenant. Example: Jurassic Park
- `resolve_full_preference_settings` (`boolean`, optional): When true, merges environment-level default preferences into the tenant's `settings.preference_set` field before returning the response. Defaults to false. Example: false
- `settings` (`object`, optional): The settings for the tenant. Includes branding and preference set. Example: {"branding": {"icon_url": "https://example.com/trex_silhouette_icon.png", "logo_url": "https://example.com/amber_fossil_logo.png", "primary_color": "#DF1A22", "primary_color_contrast": "#FFDE00"}}

Example input: `{"id":"<id>"}`

### `knock_user_channel_data_set`

Set User Channel Data · Write

Set the channel data (for example push tokens or a Slack connection) for a user on one channel.
Returns an object with __typename, channel_id, data, provider.
Needs user_id from knock_users_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `channel_id` (`string`, required): The unique identifier for the channel. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `data` (`object`, required): Channel data for a given channel type. Example: {"tokens": ["fcm-token-abc123"]}
- `user_id` (`string`, required): The unique identifier of the user. Example: user-123

Example input: `{"channel_id":"<channel_id>","data":{"tokens":["fcm-token-abc123"]},"user_id":"<user_id>"}`

### `knock_user_identify`

Identify User · Write

Create or update a user (an upsert) with properties such as name, email, phone_number, timezone and custom fields.
Returns an object with __typename, avatar, created_at, email, id, name, phone_number, timezone, updated_at.
Needs user_id from knock_users_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The unique identifier of the user. Example: user-123
- `avatar` (`string`, optional): A URL for the avatar of the user. Example: https://example.com/avatars/jane.png
- `channel_data` (`object`, optional): Channel-specific information that's needed to deliver a notification to an end provider. Example: {"97c5837d-c65c-4d54-aa39-080eeb81c69d": {"tokens": ["push_token_123"]}}
- `created_at` (`string`, optional): The creation date of the user from your system. Example: 2026-01-15T09:30:00Z
- `email` (`string`, optional): The primary email address for the user. Example: ian.malcolm@chaos.theory
- `locale` (`string`, optional): The locale of the user. Used for message localization. Example: en-US
- `name` (`string`, optional): Display name of the user. Example: Dr. Ian Malcolm
- `phone_number` (`string`, optional): The E.164 phone number of the user (required for SMS channels). Example: +14155550123
- `preferences` (`object`, optional): A set of preferences for the user. Example: {"default": {"channel_types": {"email": true}, "workflows": {"dinosaurs-loose": {"channel_types": {"email": true}}}}}
- `timezone` (`string`, optional): The timezone of the user. Must be a valid tz database time zone string. Used for recurring schedules. Example: America/New_York

Example input: `{"user_id":"<user_id>"}`

### `knock_user_merge`

Merge users · Write

Merge two users together, where the user specified with the `from_user_id` param will be merged into the user specified by `user_id`.
Returns an object with __typename, avatar, created_at, email, id, name, phone_number, timezone, updated_at.
Needs user_id from knock_users_list.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `from_user_id` (`string`, required): The user ID to merge from. Example: user_1
- `user_id` (`string`, required): The id of the user to merge into. Example: user-123

Example input: `{"from_user_id":"<from_user_id>","user_id":"<user_id>"}`

### `knock_user_preference_set_update`

Update User Preference Set · Write

Update a complete preference set for a user. By default, this is a destructive operation and will replace any existing preferences with the preferences given.
Returns an object with categories, channel_types, channels, commercial_subscribed, id, workflows.
The request replaces the stored value with what you send.
Needs user_id from knock_users_list; id from knock_user_preference_sets_list.

Read only: no. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): Unique identifier for the preference set. Example: default
- `user_id` (`string`, required): The unique identifier of the user. Example: user-123
- `__persistence_strategy__` (`string`, optional): Controls how the preference set is persisted. 'replace' will completely replace the preference set, 'merge' will merge with existing preferences. Example: merge. One of: `merge`, `replace`.
- `categories` (`object`, optional): An object where the key is the category and the values are the preference settings for that category. Example: {"marketing": false, "transactional": {"channel_types": {"email": false}}}
- `channel_types` (`object`, optional): An object where the key is the channel type and the values are the preference settings for that channel type. Example: {"email": true}
- `channels` (`object`, optional): An object where the key is the channel ID and the values are the preference settings for that channel ID. Example: {"2f641633-95d3-4555-9222-9f1eb7888a80": {"conditions": [{"argument": "US", "operator": "equal_to", "variable": "recipient.country_code"}]}, "aef6e715-df82-4ab6-b61e-b743e249f7b6": true}
- `commercial_subscribed` (`boolean`, optional): Whether the recipient is subscribed to commercial communications. When false, the recipient will not receive commercial workflow notifications. Can also be set to a settings object with conditions that are evaluated at notification send time. Example: true
- `workflows` (`object`, optional): An object where the key is the workflow key and the values are the preference settings for that workflow. Example: {"dinosaurs-loose": {"channel_types": {"email": false}}}

Example input: `{"id":"<id>","user_id":"<user_id>"}`

### `knock_users_bulk_identify`

Bulk Identify Users · Write

Identify multiple users in a single operation. Allows creating or updating up to 1,000 users in a single batch with various properties, preferences, and channel data.
Returns an object with __typename, completed_at, error_count, error_items, estimated_total_rows, failed_at, id, inserted_at, name, processed_rows, progress_path, started_at and more.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `users` (`array`, required): A list of users. Example: [{"email": "jane@ingen.net", "id": "user_1", "name": "Jane Doe", "timezone": "America/New_York"}]

Example input: `{"users":[{"email":"jane@ingen.net","id":"user_1","name":"Jane Doe","timezone":"America/New_York"}]}`

### `knock_workflow_trigger`

Trigger Workflow · Write

Trigger a Knock workflow by its key for one or more recipients (user IDs, inline user objects or object references), with optional actor, tenant, cancellation key and data for the templates.
Returns workflow_run_id for the triggered run.
Use this to send a notification through Knock. Use knock_workflow_cancel with the same cancellation_key to stop it before it finishes, and knock_schedules_create to run it on a schedule.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `key` (`string`, required): Key of the workflow to trigger. Example: new-comment
- `recipients` (`array`, required): The recipients to trigger the workflow for. Can inline identify users, objects, or use a list of user IDs. Limited to 1,000 recipients. Example: ["dr_grant", "dr_sattler", "dr_malcolm"]
- `actor` (`object`, optional): A map of properties describing a user or an object to identify in Knock and mark as who or what performed the action. Example: mr_dna
- `cancellation_key` (`string`, optional): A key that is used to reference a specific workflow trigger request when issuing a workflow cancellation request. Must be provided while triggering a workflow in order to enable subsequent cancellation. Should be unique across trigger requests to avoid unintentional cancellations. Example: isla_nublar_incident_1993
- `data` (`object`, optional): An optional map of data to pass into the workflow execution. There is a 10MB limit on the size of the full `data` payload. Any individual string value greater than 1024 bytes in length will be truncated in your logs. Example: {"order_id": "A-1001", "comment": "Looks good"}
- `settings` (`object`, optional): Optional settings that control how this workflow trigger is executed. Example: {"sandbox_mode": true, "skip_delay": true}
- `tenant` (`object`, optional): The tenant to trigger the workflow for. Triggering with a tenant will use any tenant-level overrides associated with the tenant object, and all messages produced from workflow runs will be tagged with the tenant. Example: acme-corp

Example input: `{"key":"<key>","recipients":["dr_grant","dr_sattler","dr_malcolm"]}`

### `knock_audience_members_remove`

Remove Audience Members · Destructive

Remove one or more members from the specified audience.
Returns an empty response on success (HTTP 204).

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `key` (`string`, required): The key of the audience. Example: new-comment
- `members` (`array`, required): A list of audience members to remove. You can remove up to 1,000 members per request. Example: [{"tenant": "ingen_isla_nublar", "user": {"email": "ellie@ingen.net", "id": "dr_sattler", "name": "Dr. Ellie Sattler"}}]

Example input: `{"key":"<key>","members":[{"tenant":"ingen_isla_nublar","user":{"email":"ellie@ingen.net","id":"dr_sattler","name":"Dr. Ellie Sattler"}}]}`

### `knock_object_delete`

Delete an object · Destructive

Permanently removes an object from the specified collection.
Returns an empty response on success (HTTP 204).
Needs id from knock_objects_list.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `id` (`string`, required): Unique identifier for the object. Example: project-123

Example input: `{"collection":"<collection>","id":"<id>"}`

### `knock_object_subscriptions_delete`

Delete subscriptions · Destructive

Delete subscriptions for the specified recipients from an object.
Returns an array of items with __typename, inserted_at, object, properties, recipient, updated_at.
Needs object_id from knock_objects_list.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `collection` (`string`, required): The collection this object belongs to. Example: projects
- `object_id` (`string`, required): Unique identifier for the object. Example: project-123
- `recipients` (`array`, required): The recipients of the subscription. You can subscribe up to 100 recipients to an object at a time. Example: ["user_123"]

Example input: `{"collection":"<collection>","object_id":"<object_id>","recipients":["user_123"]}`

### `knock_schedules_delete`

Delete Schedules · Destructive

Permanently deletes one or more schedules identified by the provided schedule IDs.
Returns an array of items with __typename, actor, data, id, inserted_at, last_occurrence_at, next_occurrence_at, recipient, repeats, tenant.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `schedule_ids` (`array`, required): A list of schedule IDs. Example: ["123e4567-e89b-12d3-a456-426614174000"]

Example input: `{"schedule_ids":["123e4567-e89b-12d3-a456-426614174000"]}`

### `knock_tenant_delete`

Delete a tenant · Destructive

Delete a tenant and all associated data. This operation cannot be undone.
Returns an empty response on success (HTTP 204).
Needs id from knock_tenants_list.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): The unique identifier for the tenant. Example: acme-corp

Example input: `{"id":"<id>"}`

### `knock_user_channel_data_unset`

Unset User Channel Data · Destructive

Delete channel data for a specific user and channel ID.
Returns an empty response on success (HTTP 204).
Needs user_id from knock_users_list.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `channel_id` (`string`, required): The unique identifier for the channel. Example: 4c2f0a3e-8b1d-4e5f-9a6b-7c8d9e0f1a2b
- `user_id` (`string`, required): The unique identifier of the user. Example: user-123

Example input: `{"channel_id":"<channel_id>","user_id":"<user_id>"}`

### `knock_user_delete`

Delete user · Destructive

Permanently delete a user and all associated data.
Returns an empty response on success (HTTP 204).
Needs user_id from knock_users_list.

Read only: no. Destructive: yes. Idempotent (safe to retry): yes.

Inputs:

- `user_id` (`string`, required): The ID of the user to delete. Example: user-123

Example input: `{"user_id":"<user_id>"}`

### `knock_users_bulk_delete`

Bulk Delete Users · Destructive

Permanently deletes up to 1,000 users at a time.
Returns an object with __typename, completed_at, error_count, error_items, estimated_total_rows, failed_at, id, inserted_at, name, processed_rows, progress_path, started_at and more.

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `user_ids` (`array`, required): A list of user IDs. Example: ["user_1", "user_2"]

Example input: `{"user_ids":["user_1","user_2"]}`

### `knock_workflow_cancel`

Cancel Workflow Run · Destructive

Cancel in-flight runs of a workflow that were triggered with a cancellation_key, optionally only for some recipients.
Returns an empty response on success (HTTP 204).
Use this to stop delayed, batched or multi-step workflow runs. Messages already delivered are not recalled.

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `cancellation_key` (`string`, required): A key that is used to reference a specific workflow trigger request when issuing a workflow cancellation request. Must be provided while triggering a workflow in order to enable subsequent cancellation. Should be unique across trigger requests to avoid unintentional cancellations. Example: cancel-workflow-123
- `key` (`string`, required): The key of the workflow to cancel. Example: new-comment
- `recipients` (`array`, optional): A list of recipients to cancel the notification for. If omitted, cancels for all recipients associated with the cancellation key. Example: ["jhammond"]

Example input: `{"cancellation_key":"<cancellation_key>","key":"<key>"}`

## Related

Other Communication connectors ([all 132](/agentkit/connectors/?category=communication)):

- [Gmail](/agentkit/connectors/gmail/): Scalekit connector, OAuth, 48 tools
- [Slack](/agentkit/connectors/slack/): Scalekit connector, OAuth, 96 tools
- [Google Calendar](/agentkit/connectors/googlecalendar/): Scalekit connector, OAuth, 30 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
