> **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 the Missive MCP server

The Missive MCP connector routes your AI agent's tool calls to Missive's own MCP server through Scalekit. Each user signs in to Missive once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 20 tools.

**Authentication:** OAuth 2.1/DCR
**Categories:** Communication, Collaboration, Productivity
**Tools:** 20: 13 read, 1 write, 6 destructive
**Users sign in with:** OAuth
**OAuth app:** Your own Missive MCP server app
**Built by:** Vendor MCP
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=MISSIVEMCP)

## 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 Missive MCP connection

   In **AgentKit > Connections**, create a Missive MCP connection and copy its redirect URI. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Register an OAuth app

   Missive MCP server connections use your own OAuth app. Register one with Missive MCP server and add the redirect URI you copied.

   Then enter the app's Client ID and Client Secret on the Missive MCP connection.

5. ### 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 = 'missivemcp'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Missive MCP:', 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: 'missivemcp_get_contact_details',
     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 = "missivemcp"
   identifier = "user_123"

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

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

   ```bash
   python quickstart.py
   ```

   Each user signs in once. 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: 'missivemcp_get_calendar_events',
  toolInput: {
    end_date: '<end_date>',
    start_date: '<start_date>',
  },
  connector: 'missivemcp',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="missivemcp_get_calendar_events",
    tool_input={
        "end_date": "<end_date>",
        "start_date": "<start_date>",
    },
    connection_name="missivemcp",
    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.

### `missivemcp_get_calendar_events`

Get Calendar Events · Read-only

Fetch calendar events to check the user's availability. Use this tool when: - The conversation involves scheduling, meetings, or availability; - Someone asks about meeting times or when the user is free; - Drafting a reply that proposes meeting times; If calendar_ids is omitted, queries all accessible calendars. Use mode='availability' to get just busy time slots, or mode='full' for event details. IMPORTANT: When inferring dates from conversation context: - If a month is mentioned without a year, use the NEXT occurrence of that month (future, not past); - "Upcoming", "next", or future-tense language always means a date after today; - Check the current date before making assumptions about which year.

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

Inputs:

- `end_date` (`string`, required): End date in YYYY-MM-DD format
- `start_date` (`string`, required): Start date in YYYY-MM-DD format
- `calendar_ids` (`array`, optional): Array of calendar UUIDs to query. Omit to query all accessible calendars.
- `mode` (`string`, optional): 'full' (default) returns event details, 'availability' returns only busy time slots
- `query` (`string`, optional): Search query string (full mode only)

Example input: `{"end_date":"<end_date>","start_date":"<start_date>"}`

### `missivemcp_get_contact_details`

Get Contact Details · Read-only

Look up contact details from contact books. Search by email, phone number, or general query (name, company, etc.). Returns up to 5 matching contacts.

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

Inputs:

- `email` (`string`, optional): Exact email address
- `phone_number` (`string`, optional): Phone number (digits only, e.g. 4187171866)
- `query` (`string`, optional): Name, company name, or any keyword to search

Example input: `{}`

### `missivemcp_get_conversation_entries`

Get Conversation Entries · Read-only

Fetch older entries for a conversation that was cropped. Only call this when `next_entries_before` is present and you need more history.

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

Inputs:

- `before` (`string`, required): Pass the `next_entries_before` value from a cropped conversation to fetch older entries
- `conversation_id` (`string`, required): Conversation ID to fetch entries for
- `include_attachments` (`boolean`, optional): Include file attachments for each message, internal comment and post. Supported attachments are provided as files directly; oversized or unsupported attachments include metadata and a temporary download URL only.

Example input: `{"before":"<before>","conversation_id":"<conversation_id>"}`

### `missivemcp_get_conversations`

Get Conversations · Read-only

Fetch conversations by conversation IDs or shared label IDs. Only use this if you don't have a search term, otherwise use `search_conversations` to get the conversation IDs first. One of `conversation_ids`, `shared_label_ids`, `team_ids`, `email_account_ids`, or `generic_account_ids` is required. <conversation> headers and metadata are always included. Always assess first whether you need the full content or only the headers and metadata, and pass `include_content` accordingly. If you're asked to get a conversation in a specific mailbox, always use `mailbox_id` parameter. `total_count` and `unseen_count` reflect the number of conversations on all pages. If you need file attachments to get the context, reload the conversation with `include_attachments` parameter.

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

Inputs:

- `include_content` (`boolean`, required): Fetch full content. Pass `false` if you know you only need headers and metadata.
- `after` (`string`, optional): Only fetch conversations modified after this timestamp
- `before` (`string`, optional): Only fetch conversations modified before this timestamp
- `conversation_ids` (`array`, optional): Array of conversation IDs to fetch (max 25)
- `cursor` (`string`, optional): Cursor to fetch conversations after. If `cursor` is provided, all parameters except `limit` and
`include_content` and `include_attachments` are ignored and are repeated from the previous request.
- `email_account_ids` (`array`, optional): Filter by email account IDs.
- `generic_account_ids` (`array`, optional): Filter by generic account IDs.
- `include_attachments` (`boolean`, optional): Include file attachments for each message, internal comment and post. Supported attachments are provided as files directly; oversized or unsupported attachments include metadata and a temporary download URL only.
- `limit` (`integer`, optional): Maximum number of conversations to fetch, max and default is 25
- `mailbox_id` (`string`, optional): Filter by mailbox ID. Only pass the ID from the mailbox record, otherwise use `search_mailboxes` tool first. For a team inbox, pass the mailbox ID from the team record.
- `shared_label_ids` (`array`, optional): Retrieve conversations with any of the specified shared label IDs (max 100). Only pass IDs from shared label records, otherwise use `search_shared_labels` tool first. For an email label (a mailbox ID that starts with `label-`), use `mailbox_id` instead.
- `team_ids` (`array`, optional): Retrieve conversations assigned to any of the specified team IDs.
- `unseen` (`boolean`, optional): Set to `true` to only fetch unseen.

Example input: `{"include_content":true}`

### `missivemcp_get_search_results`

Get Search Results · Read-only

Poll for additional results from a previously started conversation search. Use this after search_conversations returns partial results (done="false"). Pass the search_id from the previous response.

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

Inputs:

- `search_id` (`string`, required): UUID from previous search_conversations call

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

### `missivemcp_list_calendars`

List Calendars · Read-only

List all calendars the user has access to. Returns calendar names, IDs, and ownership status (owned vs delegated/shared). Use this to discover which calendars to query for events.

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

Inputs: none.

Example input: `{}`

### `missivemcp_list_organizations`

List Organizations · Read-only

List the organizations the user is a member of.

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

Inputs: none.

Example input: `{}`

### `missivemcp_search_conversations`

Search Conversations · Read-only

Search conversations across all connected accounts (email, chat, etc.).

NEVER use this tool if the user didn't provide a search term string, use `get_conversations` instead.
Supports filtering by email filters, internal/external scopes, and shared labels.
If you need then to get full conversation content, use `get_conversations` with the conversation IDs.

CRITICAL: This is KEYWORD search, NOT semantic search.
Multi-word queries require ALL words to appear in the email - this usually FAILS.

QUERY STRATEGY:
1. Extract the CORE searchable noun from the user's request
2. IGNORE qualifiers: "my", "last", "recent", "the", "about", "for" - these hurt the search
3. Use FILTERS (date, email) to narrow - not more keywords
4. If no/wrong results, expand to SYNONYMS (see below)

SYNONYM EXPANSION - Start with most likely term, expand if no/wrong results:
- invoice → receipt → bill → statement
- quote → estimate → proposal → pricing
- shipping → delivery → tracking → shipment
- contract → agreement → terms
- payment → paid → transaction → charge
- order → purchase → confirmation

EXAMPLES - User request → Correct approach:
  "Find my last utility invoice" → Start: query: "invoice" + since filter
    → No/wrong results? Try "bill", then "statement"
  "What was the contractor's quote?" → Start: query: "quote"
    → No/wrong results? Try "estimate", then "proposal"

WRONG: query: "last utility invoice" (0 results - emails don't contain "last")
RIGHT: query: "invoice" + filters → no/wrong results → try "bill" → try "statement"

AVAILABLE FILTERS:
- `email`: Conversations involving this address (sender OR recipient)
- `from`/`to`: Specific email header fields
- `since`/`before`: Date range (YYYY-MM-DD)
- `has_attachments`: Only with attachments
- `search_in_internal`: Search comments/posts/drafts only
- `search_in_external`: Search emails/messages only
- `mailbox_id`: Search conversations in the specified mailbox. Only pass the mailbox ID from
  the mailbox or team record, otherwise use `search_mailboxes` tool first.
  If you're asked to get a conversation in a specific mailbox, always use this parameter.

Note: A conversation can have both internal comments AND external emails/messages.

WORKFLOW:
1. Search with most likely term + filters
2. If done="false" → call get_search_results with search_id for more results
3. If too many results → add email/date filters
4. If no/wrong results → try next synonym from the list above
5. Once you have results → use get_conversations to fetch full content

If you're unsure about filters, ASK the user for clarification rather than
running a broad search. A narrow search that misses some results is better
than a broad search that returns thousands.

ASYNC BEHAVIOR: Search runs across multiple sources (Gmail, Office365, etc.).
Initial results return quickly, but if done="false", call get_search_results
with the returned search_id to get additional results as they arrive.

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

Inputs:

- `query` (`string`, required): Search query string
- `before` (`string`, optional): Only before this date (YYYY-MM-DD)
- `email` (`string`, optional): Filter to conversations involving this email address
- `email_account_ids` (`array`, optional): Search in the specified email account IDs.
- `from` (`array`, optional): Search email FROM header (provider API search)
- `generic_account_ids` (`array`, optional): Search in the specified generic account IDs.
- `has_attachments` (`boolean`, optional): Search only with attachments
- `mailbox_id` (`string`, optional): Search in the specified mailbox. Only pass the ID from the mailbox record, otherwise use `search_mailboxes` tool first
- `search_in_external` (`boolean`, optional): Search only in emails/messages (customer content)
- `search_in_internal` (`boolean`, optional): Search only in comments/posts/drafts (team content)
- `shared_label_ids` (`array`, optional): Search with any of the specified shared label IDs.
- `since` (`string`, optional): Only after this date (YYYY-MM-DD)
- `team_ids` (`array`, optional): Search in any of the specified team IDs.
- `to` (`array`, optional): Search email TO header (provider API search)
- `unseen` (`boolean`, optional): Set to `true` to only fetch unseen.

Example input: `{"query":"refund policy"}`

### `missivemcp_search_mailboxes`

Search Mailboxes · Read-only

Search unified and label mailboxes. It does not search conversations within these mailboxes, only mailboxes themselves. Does not return individual team mailboxes, retrieve them through a team record if needed. One of `query` or `types` is required. When `query` is provided, `types` is forced to "label" (to search only non-system folders). If you need to find an email account, use the `email_account_query` parameter with `types: ['all']`

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

Inputs:

- `email_account_ids` (`array`, optional): Search mailboxes in the specified email account IDs
- `email_account_query` (`string`, optional): Search mailboxes in the specified email account by username
- `limit` (`integer`, optional): Maximum number of labels to retrieve. From 1 to 100
- `only_my_accounts` (`boolean`, optional): Only return mailboxes from the user's own email accounts
- `query` (`string`, optional): Search query string
- `types` (`array`, optional): Retrieve mailboxes by type. One of: ["inbox", "archive", "flagged", "drafts", "sent", "trash", "junk", "drafts_by_me", "sent_by_me", "send_later", "commented", "closed", "snoozed", "pinned", "assigned", "assigned_to_others", "unassigned", "all", "team_unassigned"].

Example input: `{}`

### `missivemcp_search_responses`

Search Responses · Read-only

Search canned responses by semantic similarity. Use this to find relevant templates, policies, or pre-written replies that can help answer a question.

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

Inputs:

- `query` (`string`, required): Search query describing what kind of response you need
(e.g., 'refund policy', 'shipping information')

Example input: `{"query":"refund policy"}`

### `missivemcp_search_shared_labels`

Search Shared Labels · Read-only

Search shared labels across all user organizations. At least one of `query`, `label_ids`, or `parent_label_ids` is required.

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

Inputs:

- `label_ids` (`array`, optional): Retrieve labels by ID
- `limit` (`integer`, optional): Maximum number of labels to retrieve. From 1 to 100
- `organization_ids` (`array`, optional): Array of organization IDs to search in
- `parent_label_ids` (`array`, optional): Retrieve labels by parent ID
- `query` (`string`, optional): Search query string

Example input: `{}`

### `missivemcp_search_teams`

Search Teams · Read-only

Search teams across all user organizations. At least one of `query`, `team_ids`, or `organization_ids` is required.

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

Inputs:

- `any_team` (`boolean`, optional): All teams combined. If provided, other parameters are ignored. Shortcut for `query=*`
- `limit` (`integer`, optional): Maximum number of teams to retrieve. From 1 to 100
- `organization_ids` (`array`, optional): Array of organization IDs to search in
- `query` (`string`, optional): Search query string
- `team_ids` (`array`, optional): Retrieve teams by ID

Example input: `{}`

### `missivemcp_search_users`

Search Users · Read-only

Search members of the user's organizations by name or email, or retrieve them by ID, e.g. to find assignees. The user making the request is marked with `current_user`. At least one of `query`, `user_ids`, `organization_ids` or `current_user` is required.

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

Inputs:

- `current_user` (`boolean`, optional): Pass true to retrieve the user making the request ("me"). Can't be combined with other filters
- `limit` (`integer`, optional): Maximum number of users to retrieve. From 1 to 100
- `organization_ids` (`array`, optional): Only retrieve members of these organizations, from `list_organizations`
- `query` (`string`, optional): Name or email to search for
- `user_ids` (`array`, optional): Retrieve users by ID

Example input: `{}`

### `missivemcp_create_post`

Create Post · Write

Post an internal note in a conversation, e.g. to leave context or a request for teammates. The note is only visible to users with access to the conversation, it is never sent to external recipients. If you have multiple conversations to post in, do a separate tool call for each conversation.

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

Inputs:

- `conversation_id` (`string`, required): The ID of the conversation to post the note in
- `markdown` (`string`, required): The note content, in Markdown. Images are not supported and are shown as links

Example input: `{"conversation_id":"<conversation_id>","markdown":"<markdown>"}`

### `missivemcp_change_labels`

Change Labels · Destructive

Add and or remove labels from a conversation. At least one of `add_shared_label_ids`, `add_mailbox_ids`, `remove_shared_label_ids`, or `remove_mailbox_ids` is required. If you have multiple labels/mailboxes to add, do it all in a single tool call. If you have multiple conversations to change labels on, do a separate tool call for each conversation. Don't describe the labels in text.

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

Inputs:

- `conversation_id` (`string`, required): The ID of the conversation to change labels on
- `add_mailbox_ids` (`array`, optional): The IDs of the email label mailboxes to add to the conversation
- `add_shared_label_ids` (`array`, optional): The IDs of the shared labels to add to the conversation
- `remove_mailbox_ids` (`array`, optional): The IDs of the email label mailboxes to remove from the conversation.
- `remove_shared_label_ids` (`array`, optional): The IDs of the shared labels to remove from the conversation

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

### `missivemcp_compose_draft`

Compose Draft · Destructive

Create or update a message draft. Follow it with `deliver_draft` (using the returned conversation_id and draft_id) if the user asked to send, not just to draft. For email, unless the user's preferences specify otherwise: - Include a greeting like 'Hi [Name],' at the start; - Omit closings and signatures (the signature is appended automatically).

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

Inputs:

- `mode` (`string`, required): The operation to run. Each mode requires a fixed set of parameters: - new_conversation: start a new conversation; `type` sets the channel. type `email` requires `to`, `subject`, `body`. A phone provider type (whatsapp, twilio, twilio_whatsapp, signalwire, quo, dialpad) requires `to` (the recipient phone number) and `body`. - reply: reply to a message in a conversation. Requires `conversation_id`, `reply_to_id`, `body`. - new_message_in_conversation: a new, non-reply message in an existing conversation; Requires `conversation_id` and the same fields as for new_conversation. - update: change an existing draft. Requires `conversation_id`, `draft_id`, and at least one field to change. One of: `new_conversation`, `reply`, `new_message_in_conversation`, `update`.
- `bcc` (`array`, optional): Recipient email addresses (BCC). On update, omit to keep the existing recipients or pass an empty array to clear them.
- `body` (`string`, optional): The message body in plain text or Markdown. Do NOT use HTML tags. Required except an update that doesn't change the body. For email, by default include a greeting and omit closings/signatures (appended automatically); the user's preferences take precedence.
- `cc` (`array`, optional): Recipient email addresses (CC). On update, omit to keep the existing recipients or pass an empty array to clear them.
- `conversation_id` (`string`, optional): The conversation to draft in. Required for reply, new_message_in_conversation, and update.
- `draft_id` (`string`, optional): update mode: the existing draft to change.
- `from` (`string`, optional): Who to send from. Defaults to the user's default sending account.
- `reply_all` (`boolean`, optional): reply mode: include all TO and CC recipients from the message. If omitted, falls back to the user's reply-all default preference.
- `reply_to_id` (`string`, optional): reply mode: The message to reply to, ID comes from the conversation content.
- `subject` (`string`, optional): Email subject line. Required for a new email conversation.
- `to` (`array`, optional): Recipients (To). Email message: email addresses; SMS/WhatsApp: single phone number. Required for new_conversation and new_message_in_conversation. Optional for reply (overrides the derived recipients) and update (omit to keep the current recipients; email accepts an empty array to clear them).
- `type` (`string`, optional): New message type for new_conversation and new_message_in_conversation modes. Ignored by reply and update. One of: `email`, `whatsapp`, `twilio`, `twilio_whatsapp`, `signalwire`, `quo`, `dialpad`.

Example input: `{"mode":"new_conversation"}`

### `missivemcp_deliver_draft`

Deliver Draft · Destructive

Deliver a draft in the current conversation.

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

Inputs:

- `conversation_id` (`string`, required): The ID of the conversation the draft belongs to
- `draft_id` (`string`, required): The ID of the draft to deliver, from a conversation message ID or a draft creation tool response

Example input: `{"conversation_id":"<conversation_id>","draft_id":"<draft_id>"}`

### `missivemcp_manage_calendar_events`

Manage Calendar Events · Destructive

Create, update, or delete a calendar event. Use list_calendars first to discover which calendars the user has access to. Use get_calendar_events to find existing event IDs for update or delete operations. Times must be in ISO 8601 format (e.g., 2024-03-15T10:00:00). For all-day events, use date format YYYY-MM-DD. Add a video call if the user might mean it. If the video conference was created, return the link to the user.

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

Inputs:

- `action` (`string`, required): No description. One of: `create`, `update`, `delete`.
- `all_day` (`boolean`, required): Whether this is an all-day event (boolean)
- `calendar_id` (`string`, required): The UUID of the calendar to operate on. If you don't know the calendar UUID, use
list_calendars first.
- `ends_at` (`string`, required): End time in ISO 8601 format, or YYYY-MM-DD for all-day events. Don't convert timezones, just provide the time and the `timezone` param.
- `starts_at` (`string`, required): Start time in ISO 8601 format, or YYYY-MM-DD for all-day events. Don't convert timezones, just provide the time and the `timezone` param.
- `timezone` (`string`, required): Timezone for the event (e.g., 'America/New_York').
- `title` (`string`, required): The title/summary of the event
- `attendees` (`array`, optional): Array of attendee objects. Pass empty array to remove all attendees. On update, only provide
attendees array if you want to change, or empty array to remove all attendees.
- `description` (`string`, optional): The description/body of the event
- `event_id` (`string`, optional): The provider event ID. Required for update and delete. If you don't know the event ID, fetch the event first.
- `location` (`string`, optional): The location of the event
- `reminders` (`array`, optional): Array of reminder objects with minutes before the event start. Pass empty array to remove
all reminders. Google calendars support multiple reminders; Microsoft calendars use only
the first reminder. On update, only provide reminders array if you want to change, or empty
array to remove all reminders.
- `video_conference` (`boolean`, optional): Whether to attach a video conference link: Google Meet for Google calendars, Microsoft Teams for Microsoft calendars. On update: omit to keep unchanged, pass true to add, or false to remove. For Microsoft calendars, false is a no-op because online meetings cannot be removed from existing events.

Example input: `{"action":"create","all_day":true,"calendar_id":"<calendar_id>","ends_at":"<ends_at>","starts_at":"<starts_at>","timezone":"America/New_York","title":"<title>"}`

### `missivemcp_manage_tasks`

Manage Tasks · Destructive

Create a task in a conversation, or update a task's title, description, assignees, team, due date or state. Find task IDs with `get_conversations`, user IDs with `search_users` and team IDs with `search_teams`. Assignees and teams require a conversation that belongs to an organization. Do a separate tool call for each task.

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

Inputs:

- `action` (`string`, required): No description. One of: `create`, `update`.
- `conversation_id` (`string`, required): The ID of the conversation the task belongs to
- `add_assignees` (`array`, optional): The user IDs to assign the task to
- `description` (`string`, optional): The task description
- `due_at` (`string`, optional): The due date, in ISO 8601 format in the user's timezone (e.g., 2024-03-15T09:00:00)
- `remove_assignees` (`array`, optional): The user IDs to unassign from the task. Update only
- `state` (`string`, optional): The new task state. Update only. One of: `todo`, `in_progress`, `completed`.
- `task_id` (`string`, optional): The ID of the task to update. Required for update
- `team` (`string`, optional): The ID of the team of the task
- `title` (`string`, optional): The task title. Required for create

Example input: `{"action":"create","conversation_id":"<conversation_id>"}`

### `missivemcp_update_conversation`

Update Conversation · Destructive

Change a conversation's subject, color, users, assignees, team and status. All changes apply to everyone in the conversation, except `archive`, `snooze_until` and `unarchive` which only apply to the current user; snoozing also archives the conversation. Find user IDs with `search_users`, team IDs with `search_teams` and organization IDs with `list_organizations`. At least one change is required. Do all changes to a conversation in a single tool call, and a separate tool call for each conversation.

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

Inputs:

- `conversation_id` (`string`, required): The ID of the conversation to update
- `add_assignees` (`array`, optional): The user IDs to assign to the conversation
- `add_users` (`array`, optional): The user IDs to add to the conversation, without assigning them
- `archive` (`boolean`, optional): Pass `true` to archive the conversation for the current user
- `close` (`boolean`, optional): Pass `true` to close the conversation
- `color` (`string`, optional): The conversation color, as a hex color code (e.g., #3b82f6)
- `organization` (`string`, optional): The ID of the organization to move a personal conversation to. Required to add users, assignees or a team to a personal conversation
- `remove_assignees` (`array`, optional): The user IDs to unassign from the conversation. Unassigning everyone from an open team conversation moves it to the team inbox
- `reopen` (`boolean`, optional): Pass `true` to reopen the conversation
- `snooze_until` (`string`, optional): Snooze the conversation for the current user until this time, in ISO 8601 format in the user's timezone (e.g., 2024-03-15T09:00:00)
- `subject` (`string`, optional): The new conversation subject
- `team` (`string`, optional): The ID of the team to move the conversation to. The conversation also goes to the team inbox and is unassigned from everyone, unless `add_assignees` or `close` is passed
- `unarchive` (`boolean`, optional): Pass `true` to move the conversation back to the current user's inbox, ending its snooze
- `unsnooze_on_reply` (`boolean`, optional): Whether to unsnooze the conversation when a reply is received. Defaults to the user's setting in the app. Requires `snooze_until`

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

## 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 |
