Vapi
Vapi is a voice AI platform that enables developers to build, test, and deploy voice agents with natural conversations, real-time responses, and seamless integration capabilities.
Verdict
Common use cases
- Provision phone numbers for new campaigns
- Analyze call volume and duration trends
- Create voice assistants for appointment booking
- Test conversation flows before deployment
- Monitor call quality with automated alerts
Integration
- Vendor
- Vapi
- Category
- other
- Auth
- API_KEY
- Tools
- 42
- Composio slug
vapi
Tools
- Create Analytics Queries
Tool to create and execute analytics queries on VAPI data. Use when you need to analyze call or subscription metrics with aggregations like count, sum, average, min, or max. Supports grouping by various dimensions and time-based analysis wi
- Create Assistant
Tool to create a new Vapi assistant with specified transcriber, voice, and AI model configurations. Use when setting up a conversational AI assistant for voice interactions. The assistant requires transcription (speech-to-text), voice (text
- Create Eval
Tool to create an eval for testing conversation flows. Use when you need to validate that an AI assistant responds correctly to specific conversation scenarios.
- Create Monitoring Policy
Tool to create a monitoring policy in VAPI. Use when you need to set up automated monitoring rules based on thresholds and time windows. Policies can trigger alerts based on event counts or percentages over a specified lookback window.
- Create OpenAI Chat
Tool to create an OpenAI-compatible chat using the Vapi API. Use when you need to send a chat message to an assistant or squad and receive a response. Supports both streaming and non-streaming modes.
- Create Phone Number
Tool to create a phone number with VAPI. Use when you need to provision a new phone number for voice AI applications. Supports multiple providers (VAPI, Twilio, Vonage, Telnyx, BYO). Required parameters vary by provider.
- Create Phone Number
Tool to create a phone number with Vapi. Supports multiple providers including byo-phone-number, twilio, vonage, vapi, and telnyx. Use this to provision new phone numbers for handling voice calls. For vapi provider, only provider field is r
- Create Provider Resource
Tool to create an 11Labs pronunciation dictionary resource. Use when you need to define custom pronunciations for specific terms or acronyms in voice synthesis.
- Create Scorecard
Tool to create a scorecard for observability and evaluation. Use when setting up metrics to evaluate assistant performance based on structured outputs. Scorecards contain metrics with conditions that calculate normalized scores on a 100-poi
- Create Session
Tool to create a new session in Vapi. Use when you need to establish a persistent conversation context that can span multiple chats. Sessions automatically expire after 24 hours.
- Delete Calldestructive
Tool to delete a call by its unique identifier. Use when you need to remove call data from the system. Returns the deleted call object with all its associated data.
- Delete Chatdestructive
Tool to delete a chat by its ID from Vapi. Use when you need to permanently remove a chat conversation.
- Delete Evaldestructive
Tool to delete an eval by ID. Use when you need to permanently remove an eval from the system.
- Delete Eval Rundestructive
Tool to delete an eval run by its ID from Vapi. Use when you need to permanently remove an evaluation run.
- Delete Phone Numberdestructive
Tool to delete a phone number from Vapi. Use when you need to remove a phone number from your Vapi organization. Returns the deleted phone number object.
- Get Assistant
Tool to retrieve a specific assistant by ID from Vapi. Use when you need to fetch details about an existing assistant.
- Get Call
Tool to fetch call details by ID. Use when you have a call ID and need full call information including status, duration, costs, messages, and recordings.
- Get Chat
Tool to fetch chat details by ID. Use when you have a chat ID and need full chat information including messages, costs, and configuration.
- Get Eval
Tool to retrieve an eval by its ID. Use when you need to fetch details about a specific eval including its mock conversation messages and metadata.
- Get File
Tool to retrieve a file by its ID from Vapi. Use when you need to get details about a specific file including its status, metadata, storage location, and timestamps.
- Get Insights
Tool to retrieve insights from Vapi. Use when you need to fetch insight data with optional filtering by ID, timestamps, or pagination. Returns a paginated list of insights with metadata.
- Get Insights
Tool to retrieve insights from VAPI. Use when you need to list insights with optional filtering by ID, creation date, or update date. Supports pagination and sorting.
- Get Observability Scorecard
Tool to list observability scorecards with optional filtering and pagination. Use when you need to retrieve scorecards for monitoring or analysis.
- Get Tool
Tool to fetch tool details by ID. Use when you have a tool ID and need full tool configuration including type, messages, function definitions, and server settings.
- List Assistants
Tool to list all assistants in your VAPI organization. Use when you need to retrieve information about configured assistants. Supports filtering by creation and update timestamps.
- List Calls
Tool to list calls from Vapi. Use when you need to retrieve multiple calls with optional filtering by id, assistantId, phoneNumberId, or date ranges. Returns an array of call objects with details including status, costs, messages, and artif
- List Chats
Tool to retrieve a list of chat conversations from VAPI. Use when you need to view existing chats, optionally filtered by assistant, squad, session, or time range. Supports pagination and sorting for efficient retrieval of large chat histor
- List Evals
Tool to retrieve a paginated list of evals from Vapi. Use when you need to list or search evals with optional filtering by creation/update timestamps.
- List Monitoring Policies
Tool to retrieve monitoring policies from Vapi. Use when you need to list, filter, or search for monitoring policies configured in the organization. Supports filtering by policy ID, severity level, monitor ID, and date ranges for creation/u
- List Provider Resources
Tool to retrieve provider resources from Vapi (e.g., 11labs pronunciation dictionaries). Use when you need to list or search provider-specific resources with optional filtering by timestamps.
- List Provider Resources
Tool to list provider resources from Vapi. Use when you need to retrieve pronunciation dictionaries or other resources from providers like 11labs or Cartesia.
- List Scorecards
Tool to retrieve a paginated list of scorecards from Vapi. Use when you need to list or search scorecards with optional filtering by creation/update timestamps.
- List Sessions
Tool to retrieve a paginated list of sessions from VAPI. Use when you need to list sessions with optional filtering by session ID, name, assistant, squad, or workflow. Supports pagination, sorting, and timestamp-based filtering.
- List Structured Outputs
Tool to list structured outputs with optional filtering. Use when you need to retrieve structured output configurations with pagination support. Supports filtering by ID, name, timestamps, and includes pagination controls.
- List Structured Outputs
Tool to list structured outputs with optional filtering and pagination. Use when you need to retrieve structured output configurations from Vapi.
- Test Code Tool Execution
Tool to test TypeScript code execution in Vapi's code tool environment. Use when validating code before deploying it as a tool.
- Update Assistant
Tool to update an existing Vapi assistant configuration. Use when you need to modify assistant properties such as name, voice settings, transcriber configuration, model settings, messages, or other behavior parameters. Only include fields y
- Update Eval
Tool to update an existing eval in Vapi. Use when you need to modify eval properties like name, description, type, or mock conversation messages.
- Update Insight
Tool to update an existing insight configuration in VAPI. Use when you need to modify insight properties like name, queries, time range, or visualization settings. Supports all four insight types (bar, pie, line, text) with type-specific co
- Update Phone Number
Tool to update an existing phone number configuration in VAPI. Use when you need to modify phone number settings such as name, associated assistant/workflow, or provider-specific configurations.
- Update Tool
Tool to update an existing Vapi tool configuration. Use when you need to modify tool properties such as function definitions, server settings, messages, or other tool-specific parameters.
- Upload File
Tool to upload a file to Vapi Knowledge Base. Use when you need to add files for AI assistants to reference. Returns file metadata including ID, storage URLs, and processing status.
Setup
Setup guide
- 11. Open your Switchy workspace and navigate to Settings > Integrations > Add MCP. 2. Search for Vapi and click Connect. 3. Log into your Vapi dashboard at vapi.ai, navigate to API Keys under your account settings, and generate a new key with full read/write permissions. 4. Paste the API key into Switchy's connection dialog and click Authorize. 5. Return to any Space and type '@Vapi' to confirm the integration appears in the mention menu. 6. Test the connection by asking '@Vapi list my assistants' or '@Vapi show recent call analytics'. 7. If the MCP returns data, you're ready to create assistants, provision numbers, or query call metrics directly from chat.
What teammates see: by default, memories from Vapi are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Create Support Assistant
@Vapi create an assistant named 'Support Bot' using the Deepgram transcriber and ElevenLabs voice. Configure it to answer common product questions and escalate complex issues to a human agent.Open in a Space →
Analyze Last Week's Calls
@Vapi create an analytics query for the last 7 days showing total call count, average call duration, and percentage of calls that ended successfully. Group results by day.Open in a Space →
Provision Campaign Number
@Vapi provision a new phone number using Twilio in the 415 area code. Assign it to the 'Q1 Outreach' assistant and enable call recording.Open in a Space →
Test Conversation Flow
@Vapi create an eval that tests whether the 'Booking Assistant' correctly captures a customer's preferred appointment time and confirms availability for next Tuesday at 2pm.Open in a Space →
Set Call Quality Alert
@Vapi create a monitoring policy that triggers an alert if the call failure rate exceeds 5% over any 1-hour window, or if average call duration drops below 30 seconds.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@vapi show me analytics for all calls this month grouped by assistant, including average call duration and total minutes used
Here's a summary of your call analytics for this month: • Customer Support Assistant: 342 calls, avg 4.2 min, 1,436 total minutes • Sales Qualifier Bot: 189 calls, avg 2.8 min, 529 total minutes • Appointment Scheduler: 156 calls, avg 3.1 min, 484 total minutes Total across all assistants: 687 calls, 2,449 minutes consumed. Your busiest day was the 15th with 89 calls. This data reflects completed calls only; abandoned calls under 10 seconds are excluded from duration averages.
This example shows Vapi's analytics aggregation capability. The query requires your API key to have read access to call logs. Note that real-time data may lag by a few minutes, and complex queries with multiple groupings can take 5-10 seconds to execute depending on your account's call volume.
@vapi create a new assistant named 'Product Demo Bot' using GPT-4 with a professional female voice, optimized for explaining our SaaS features to inbound leads
Assistant created successfully: • Name: Product Demo Bot • ID: asst_8k3m9n2p4q • Model: GPT-4 (gpt-4-0125-preview) • Voice: ElevenLabs 'Rachel' (professional, neutral tone) • Transcriber: Deepgram Nova-2 The assistant is now active and ready to handle calls. You'll need to assign it to a phone number or embed it in your web app to start receiving interactions. Default settings include a 30-second silence timeout and standard interruption handling.
This demonstrates Vapi's assistant creation workflow. The MCP configures transcriber, model, and voice provider in one step, but you'll need separate actions to attach the assistant to a phone number or configure custom prompts. Voice selection depends on your ElevenLabs or other TTS provider credits.
@vapi provision a new US phone number in area code 415 using the Vapi provider, then attach it to our 'Product Demo Bot' assistant for immediate use
Phone number provisioned and configured: • Number: +1 (415) 555-0847 • Provider: Vapi (native) • Status: Active • Attached to: Product Demo Bot (asst_8k3m9n2p4q) The number is live and will route incoming calls to your assistant immediately. Outbound calling is enabled by default. You're billed per-minute for both inbound and outbound usage. Test the setup by calling the number — expect a 1-2 second delay before the assistant responds on first connection.
This example chains two Vapi operations: provisioning a number and linking it to an assistant. The action has side effects (you're billed for the number even if unused) and requires your API key to have write permissions. Number availability varies by area code; if 415 is exhausted, Vapi will return an error rather than auto-selecting an alternative.
Use-case deep-dives
When Vapi makes sense for high-volume phone support
A 6-person support team fielding 200+ inbound calls per day about order status, returns, and basic troubleshooting should consider Vapi if they already have structured data in a CRM or ticketing system. The MCP's phone-number provisioning and analytics-query tools let you spin up a voice assistant that handles tier-1 questions while your team focuses on escalations. The threshold: if your FAQs change weekly or your product requires nuanced judgment calls, the assistant-creation overhead (transcriber config, voice tuning, eval flows) becomes a maintenance tax. Vapi wins when your call scripts are stable and your data sources are API-accessible. If you're routing fewer than 50 calls per day, a chatbot on your site is cheaper and easier to iterate.
Using Vapi for lead-scoring calls without burning reps
A 3-person sales team with 500 inbound leads per month can use Vapi to automate the first-touch qualification call—asking budget, timeline, and decision-maker questions before a human rep gets involved. The MCP's create-assistant and create-phone-number tools let you provision a voice agent that dials leads, logs responses, and flags high-intent prospects for follow-up. The trade-off: if your ICP is complex or your pitch requires storytelling, the assistant will sound robotic and hurt conversion. Vapi works when your qualification script is a fixed set of yes/no or multiple-choice questions and you can feed the results into a CRM via webhook. If your close rate depends on rapport-building in the first 60 seconds, keep humans on the phone.
When Vapi beats survey emails for response rates
A 4-person product team shipping a consumer app with 1,000 monthly purchases can use Vapi to call customers 48 hours post-order and ask three NPS-style questions. The MCP's monitoring-policy and analytics-query tools let you track completion rates and flag sentiment drops in real time. The win: voice calls get 3-5x the response rate of email surveys, and the transcripts surface verbatim pain points your team can triage in standups. The boundary: if your customer base skews international or your product is B2B with multi-stakeholder buyers, a phone call feels intrusive and the transcription accuracy drops on accents. Vapi is the right call when your users are domestic, your questions are short, and you need qualitative signal fast.
Frequently asked
What does the Vapi MCP let me do in Switchy?
It lets you build and manage voice AI assistants directly from Switchy's chat interface. You can create assistants with custom transcribers and voices, provision phone numbers across providers like Twilio or Telnyx, run analytics queries on call data, and set up monitoring policies to track conversation metrics. Think of it as programmatic control over your entire Vapi voice infrastructure without leaving the workspace.
Do I need a Vapi account to use this MCP?
Yes. You need an active Vapi account and an API key with write permissions. The MCP authenticates using that key, so whoever connects it in Switchy must have access to your team's Vapi dashboard. If you're on a free Vapi tier, some tools like phone number provisioning may hit vendor limits before Switchy's.
Can the Vapi MCP handle live calls or only assistant setup?
It handles both setup and runtime operations. You can create assistants, provision numbers, and send chat messages to those assistants in streaming or non-streaming mode. However, it doesn't give you a real-time audio stream inside Switchy — for that, you'd still use Vapi's web SDK or phone interface. This MCP is for orchestration and analytics, not replacing the call experience itself.
How is this different from using Vapi's dashboard or API directly?
The dashboard is faster for one-off assistant tweaks; the API is better for production integrations. This MCP sits in between: it's useful when you want to prototype voice flows in natural language, run ad-hoc analytics queries without writing code, or let non-technical teammates provision test numbers. You trade Vapi's full UI polish for the speed of typing a request in Switchy.
Who on the team should connect the Vapi MCP?
Whoever owns your Vapi account and is comfortable sharing API key access. In practice, that's usually a product or ops lead who already manages voice infrastructure. Once connected, any Switchy user in the workspace can invoke the tools, so treat it like shared infra — don't connect it if you're not ready for the whole team to create test assistants or query call logs.