Connect AI agents to the Vapi MCP server

Vendor MCP

The Vapi MCP connector routes your AI agent's tool calls to Vapi's own MCP server through Scalekit. Each user connects their own Vapi access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 14 tools.

Tools: 14
What they doRead · write · destructive: 8 · 6 · 08 read6 write0 destructive
Users sign in with: Access token

What you can do

Manage voice assistants: create, update, and review voice AI assistants and their model and voice settings
Place outbound calls: start or schedule calls with a chosen assistant and phone number
Review calls: list calls and fetch a call’s status, duration, and other details
Manage assistant tools: create, update, and list tools that extend what assistants can do on calls
Look up phone numbers: list the account’s phone numbers and fetch details for a specific one

Setup

npm install @scalekit-sdk/node dotenv

pip install scalekit-sdk-python python-dotenv

Set your credentials
Add your Scalekit credentials to your .env file. Find values in app.scalekit.com > Developers > API Credentials.
.env
```
SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
SCALEKIT_CLIENT_ID=<your-client-id>
SCALEKIT_CLIENT_SECRET=<your-client-secret>
```
Create the Vapi MCP connection
In AgentKit > Connections, create a Vapi MCP connection. The name you give it is the connection_name your code passes. See Configure connections.
Console steps with screenshots
Register your Vapi API key with Scalekit so it can authenticate and proxy requests to the Vapi MCP server on behalf of your users. Vapi MCP uses bearer token authentication. Scalekit sends the key as Authorization: Bearer <token> to the Vapi MCP endpoint at https://mcp.vapi.ai/mcp.
1. Get a Vapi API key
  - Go to dashboard.vapi.ai/org/api-keys and sign in or create an account.
  - Under API Keys, copy your Private Key. Use the private key — it authorizes server-side calls to the Vapi API.
  - Treat this key as a secret. Anyone with it can manage your assistants, calls, and phone numbers.
2. Create a connection in Scalekit
  - In the Scalekit dashboard, go to AgentKit > Connections. Find Vapi MCP and click Create.
  - Note the Connection name — you use this as connection_name in your code (for example, vapimcp).
3. Add a connected account
  
  Connected accounts link a specific user identifier in your system to a Vapi API key. Add one in the dashboard to test. In production, each user adds their own through the authorization link: they enter their credentials on the page it opens.
  
  In the dashboard, to test
  - Open the connection you created and click the Connected Accounts tab → Add account.
  - Fill in:
    
    Your User’s ID — a unique identifier for this user in your system (for example, user_123)
    
    Vapi API key — the private key you copied in step 1
  - Click Save.
  From your backend, if your app already has the credentials
  
  For example, when users enter them on a settings page in your app:
  Node.js
  Python
  1 import { ConnectorStatus } from '@scalekit-sdk/node' 2 3 const authorizationDetails = { 4 details: { 5 case: 'staticAuth', 6 value: { details: { token: process.env.VAPI_API_KEY } }, 7 }, 8 }; 9 10 let { connectedAccount } = await scalekit.actions.upsertConnectedAccount({ 11 connectionName: 'vapimcp', 12 identifier: 'user_123', // your user's unique ID 13 authorizationDetails, 14 }); 15 16 // Make sure the account is ACTIVE before the first tool call. 17 if (connectedAccount?.status !== ConnectorStatus.ACTIVE) { 18 ({ connectedAccount } = await scalekit.actions.upsertConnectedAccount({ 19 connectionName: 'vapimcp', 20 identifier: 'user_123', 21 authorizationDetails, 22 })); 23 }

Authorize a user and make your first call

Node.js
Python

1
import { ScalekitClient } from '@scalekit-sdk/node'
2
import 'dotenv/config'
3
import { createInterface } from 'node:readline/promises'
4

5
const scalekit = new ScalekitClient(
6
  process.env.SCALEKIT_ENVIRONMENT_URL,
7
  process.env.SCALEKIT_CLIENT_ID,
8
  process.env.SCALEKIT_CLIENT_SECRET,
9
)
10
const actions = scalekit.actions
11

12
const connector = 'vapimcp'
13
const identifier = 'user_123'
14

15
// Generate an authorization link for the user
16
const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
17
console.log('Authorize Vapi MCP:', link)
18
const rl = createInterface({ input: process.stdin, output: process.stdout })
19
await rl.question('Press Enter after authorizing...')
20
rl.close()
21

22
// Make your first call
23
const result = await actions.executeTool({
24
  connector,
25
  identifier,
26
  toolName: 'vapimcp_list_assistants',
27
  toolInput: { context: 'Retrieving a customer\'s recent orders to investigate a billing issue and help support determine the appropriate resolution.', llm_model: 'claude-opus-4-8', rationale: 'YOUR_RATIONALE' },
28
})
29
console.log(result)

npx tsx quickstart.mts

1
import os
2
from scalekit import ScalekitClient
3
from dotenv import load_dotenv
4
load_dotenv()
5

6
scalekit_client = ScalekitClient(
7
    env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
8
    client_id=os.getenv("SCALEKIT_CLIENT_ID"),
9
    client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
10
)
11
actions = scalekit_client.actions
12

13
connection_name = "vapimcp"
14
identifier = "user_123"
15

16
# Generate an authorization link for the user
17
link_response = actions.get_authorization_link(
18
    connection_name=connection_name,
19
    identifier=identifier,
20
)
21
print("Authorize Vapi MCP:", link_response.link)
22
input("Press Enter after authorizing...")
23

24
# Make your first call
25
result = actions.execute_tool(
26
    tool_input={"context":"Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution.","llm_model":"claude-opus-4-8","rationale":"YOUR_RATIONALE"},
27
    tool_name="vapimcp_list_assistants",
28
    connection_name=connection_name,
29
    identifier=identifier,
30
)
31
print(result)

python quickstart.py

Each user opens the link once and enters their Vapi MCP server credentials there. If your app already has a user's credentials, add the account from your backend instead, as the console steps above show. See Authorize a user for the full flow and statuses.

Tools

Pass the exact name to execute_tool

Try in Playground Request a tool

vapimcp_get_assistantRetrieves the full configuration of a specific Vapi assistant by ID, including its LLM, voice, transcription settings, tools, and first message.Read-only
Get assistant
Retrieves the full configuration of a specific Vapi assistant by ID, including its LLM, voice, transcription settings, tools, and first message.
Inputs
assistantIdstringrequired
ID of the assistant to get
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_get_callRetrieves detailed information about a specific Vapi call by ID, including its status, duration, transcript, recording URL, and associated assistant.Read-only
Get call
Retrieves detailed information about a specific Vapi call by ID, including its status, duration, transcript, recording URL, and associated assistant.
Inputs
callIdstringrequired
ID of the call to get
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_get_phone_numberRetrieves details of a specific Vapi phone number by ID, including its number, provider, and any assistant or squad configuration attached to it.Read-only
Get phone number
Retrieves details of a specific Vapi phone number by ID, including its number, provider, and any assistant or squad configuration attached to it.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
phoneNumberIdstringrequired
ID of the phone number to get
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_get_toolRetrieves the full configuration of a specific Vapi tool by ID, including its type (SMS, transfer call, function, or API request) and associated settings.Read-only
Get tool
Retrieves the full configuration of a specific Vapi tool by ID, including its type (SMS, transfer call, function, or API request) and associated settings.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
toolIdstringrequired
ID of the tool to get
vapimcp_list_assistantsLists all Vapi assistants configured in the account.Read-only
List assistants
Lists all Vapi assistants configured in the account. Returns a list of assistant objects including their IDs, names, configurations, and settings.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_list_callsLists all Vapi calls in the account.Read-only
List calls
Lists all Vapi calls in the account. Returns call records including their IDs, status, duration, associated assistant, and transcript metadata.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_list_phone_numbersLists all phone numbers provisioned in the Vapi account.Read-only
List phone numbers
Lists all phone numbers provisioned in the Vapi account. Returns phone number objects including their IDs, numbers, provider, and associated assistant configurations.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_list_toolsLists all Vapi tools configured in the account.Read-only
List tools
Lists all Vapi tools configured in the account. Vapi tools extend assistant capabilities — types include SMS, transfer call, custom functions, and API request tools.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
vapimcp_create_assistantCreates a new Vapi voice AI assistant with the specified configuration.Write
Create assistant
Creates a new Vapi voice AI assistant with the specified configuration. Configure the assistant's LLM provider, voice, transcription engine, first message, and any tools it should have access to.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
namestringrequired
Name of the assistant
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
firstMessagestring
First message to say to the userdefault Hello, how can I help you today?
firstMessageModestring
This determines who speaks first, either assistant or userone of assistant-speaks-firstassistant-waits-for-userassistant-speaks-first-with-model-generated-messagedefault assistant-speaks-first
instructionsstring
Instructions for the assistantdefault You are a helpful assistant.
llmstring
LLM configuration
toolIdsarray
IDs of tools to use with this assistant
transcriberobject
Transcription configuration
voiceobject
Voice configuration
vapimcp_create_callInitiates or schedules an outbound Vapi call.Write
Create call
Initiates or schedules an outbound Vapi call. Specify the assistant and phone number to use, the customer's phone number, and optionally a scheduled time. Use assistantOverrides.variableValues to inject dynamic data into the assistant's prompts.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
assistantIdstring
ID of the assistant to use for the call
assistantOverridesobject
Overrides for the assistant configuration
customerobject
Customer information
phoneNumberIdstring
ID of the phone number to use for the call
scheduledAtstring
ISO datetime string for when the call should be scheduled (e.g. "2025-03-25T22:39:27.771Z")
vapimcp_create_toolCreates a new Vapi tool that can be attached to assistants.Write
Create tool
Creates a new Vapi tool that can be attached to assistants. Supports four tool types: 'sms' for sending text messages, 'transferCall' for transferring calls to destinations, 'function' for custom server-side functions, and 'apiRequest' for HTTP API integrations.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
typestringrequired
Type of the tool to createone of smstransferCallfunctionapiRequest
apiRequestobject
API Request tool configuration - for HTTP API integration
descriptionstring
Description of what the function/tool does
functionobject
Custom function tool configuration - for custom server-side functions
namestring
Name of the function/tool
smsobject
SMS tool configuration - to send text messages
transferCallobject
Transfer call tool configuration - to transfer calls to destinations
vapimcp_send_feedbackSend feedback about this server to its developers.Write
Send Feedback
Send feedback about this server to its developers. Most important: report a missing capability whenever no available tool fits your task, even if you can work around it (feedback_type "missing_capability"). Also welcome: a tool that failed or confused you, an unhelpful error, or something that worked well. This records the feedback; it does not add or change tools. Do not include user PII or sensitive content.
Inputs
feedback_typestringrequired
What kind of feedback this is. Use 'missing_capability' when the tool you needed does not exist in the tool list - nothing failed, the capability is absent (this is the most valuable report; send it even if you found a workaround). Use 'issue' when an existing tool behaved badly: it failed, returned a confusing error, or its description or schema misled you. Use 'praise' when something worked notably well. Use 'other' for anything else.one of missing_capabilityissuepraiseother
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
summarystringrequired
One self-contained sentence. For 'missing_capability': the capability you needed, e.g. 'No tool to delete multiple cohorts in one call.' For 'issue': the tool and the problem, e.g. 'query-trends rejects relative date ranges with an unclear error.'
detailsstring
Optional longer context: what you tried, exact parameter values, the error text you saw, and any workaround you used. Omit when the summary says it all.
friction_pointsstring
Optional: the specific moments that slowed you down, as short bullet-like sentences, quoting exact tool names, parameters, or error text.
sentimentstring
Optional: how the experience felt overall.one of positiveneutralnegativemixed
suggested_improvementstring
Optional: the concrete change that would have helped, e.g. the tool to add, the description to reword, or the error message to improve.
task_completedboolean
Optional: whether you still completed the user's task despite the problem.
tool_namestring
Optional: the existing tool this feedback is about (for 'issue' or 'praise'). Leave empty for 'missing_capability' - the point is that no tool fits.
vapimcp_update_assistantUpdates an existing Vapi assistant's configuration.Write
Update assistant
Updates an existing Vapi assistant's configuration. Only fields you provide will be changed; omitted fields retain their current values.
Inputs
assistantIdstringrequired
ID of the assistant to update
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
firstMessagestring
First message to say to the user
firstMessageModestring
This determines who speaks first, either assistant or userone of assistant-speaks-firstassistant-waits-for-userassistant-speaks-first-with-model-generated-message
instructionsstring
New instructions for the assistant
llmstring
New LLM configuration
namestring
New name for the assistant
toolIdsarray
New IDs of tools to use with this assistant
transcriberobject
New transcription configuration
voiceobject
New voice configuration
vapimcp_update_toolUpdates an existing Vapi tool's configuration.Write
Update tool
Updates an existing Vapi tool's configuration. Only fields you provide will be changed. Supports all tool types: sms, transferCall, function, and apiRequest.
Inputs
contextstringrequired
Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution."
llm_modelstringrequired
The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" â never guess.
rationalestringrequired
Always provide a brief explanation of why you are calling this tool
toolIdstringrequired
ID of the tool to update
apiRequestobject
API Request tool configuration - for HTTP API integration
descriptionstring
Description of what the function/tool does
functionobject
Custom function tool configuration - for custom server-side functions
namestring
Name of the function/tool
smsobject
SMS tool configuration - to send text messages
transferCallobject
Transfer call tool configuration - to transfer calls to destinations

Workflows

Tool calling

Use vapimcp_list_assistants to discover existing assistants, then vapimcp_get_assistant to read a specific assistant’s full configuration before changing it.
Use vapimcp_create_assistant to provision a new voice AI assistant, and vapimcp_update_assistant to change only the fields you pass — omitted fields keep their current values.
Use vapimcp_list_phone_numbers and vapimcp_get_phone_number to find the phone number to call from.
Use vapimcp_create_call to place an outbound call with a chosen assistant and phone number, then vapimcp_get_call to read its status, duration, and transcript.
Use vapimcp_create_tool to add SMS, transfer-call, function, or API-request tools, and attach them to an assistant with vapimcp_update_assistant.

Node.js
Python

const toolResponse = await actions.executeTool({
  connector: 'vapimcp',
  identifier: 'user_123',
  toolName: 'vapimcp_create_call',
  toolInput: {
    assistantId: 'YOUR_ASSISTANT_ID',
    phoneNumberId: 'YOUR_PHONE_NUMBER_ID',
    customer: { number: '+15551234567' },
  },
});
console.log('Call started:', toolResponse.data);

tool_response = actions.execute_tool(
    connection_name='vapimcp',
    identifier='user_123',
    tool_name='vapimcp_create_call',
    tool_input={
        "assistantId": "YOUR_ASSISTANT_ID",
        "phoneNumberId": "YOUR_PHONE_NUMBER_ID",
        "customer": {"number": "+15551234567"},
    },
)
print("Call started:", tool_response)

What you can do

Setup

Install the SDK

Set your credentials

Create the Vapi MCP connection

Get a Vapi API key

Create a connection in Scalekit

Add a connected account

Authorize a user and make your first call

Tools

Call it

Returns

Get assistant

Inputs

Get call

Inputs

Get phone number

Inputs

Get tool

Inputs

List assistants

Inputs

List calls

Inputs

List phone numbers

Inputs

List tools

Inputs

Create assistant

Inputs

Create call

Inputs

Create tool

Inputs

Send Feedback

Inputs

Update assistant

Inputs

Update tool

Inputs

Workflows

Related: other Communication connectors