Find campaign failures by tracing a single phone call. 
Claude 
ChatGPT 
Cursor 
Gemini 
Windsurf 
VS Code 
JetBrains 
Vercel
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Marchex MCP Server connects your AI client to Marchex's conversation intelligence platform. Use it to search call records, analyze performance metrics, and manage campaign data.
Your agent can list accounts, find specific calls using `search_calls`, retrieve granular call details via `get_call_details`, and calculate overall campaign health with `get_call_analytics`.
What your AI can do
Get account details
Retrieves detailed information for a single specified user or organizational account.
Get call analytics
Calculates and returns aggregated call performance metrics across a given time period.
Get call details
Fetches all available metadata for one specific, identified phone call record.
Find specific call instances using timeframes or phone numbers with search_calls, then fetch complete metadata on any single call via get_call_details.
List all active campaigns (list_campaigns) and run deep analyses to get aggregated performance metrics using get_call_analytics.
See what numbers are available with list_numbers, check specific account details with get_account_details, or list all user accounts with list_accounts.
View the setup and configuration of any tracking campaign using get_campaign_details.
Get a list of all registered users (list_users) or check details on specific tracking phone numbers with get_number_details.
Ask an AI about this
Compatible AI Apps
OAuth 2.0 Compatible Gemini Waiting for input…
Marchex MCP Server: 10 Tools for Conversation Intelligence
Use these tools in your agent to get granular access to Marchex data. You can list accounts, search calls, retrieve analytics, and manage campaign configurations all from one chat interface.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Marchex on VinkiusGet Account Details
Retrieves detailed information for a single specified user or organizational account.
Get Call Analytics
Calculates and returns aggregated call performance metrics across a given time...
Get Call Details
Fetches all available metadata for one specific, identified phone call record.
Get Campaign Details
Retrieves the full configuration and status details for a single tracking campaign.
Get Number Details
Gets specific operational information about one registered phone number.
List Accounts
Returns a list of all user accounts associated with the Marchex system.
List Campaigns
Lists every active and inactive tracking campaign setup in your account.
List Numbers
Returns a list of all phone numbers that can be used for tracking calls.
List Users
Generates a simple listing of all user accounts associated with the platform.
Search Calls
Searches through your entire call history using criteria like date range or phone...
Connect to your AI in seconds. Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Marchex integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Marchex, then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Marchex. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 10 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Trying to figure out why a campaign failed always feels like digging through old spreadsheets.
Today, diagnosing a failure means jumping into the Marchex dashboard. You pull up the 'Campaign Performance' tab, get a vague number for 'calls missed,' then you copy that ID over to a spreadsheet to cross-reference it with call records from another system, hoping the fields align.
With this MCP server, your agent handles all those jumps. Instead of copying and pasting data between three different views, you just ask: 'Why did Campaign X fail?' The agent runs `get_campaign_details`, finds the necessary IDs, executes `search_calls`, and gives you a direct answer.
Marchex MCP Server lets you get precise call metrics.
You used to have to run separate reports for 'total calls' versus 'average duration.' This meant running two distinct queries, manually comparing the results in a spreadsheet, and risking mismatched date ranges every time.
Now, you ask your agent to calculate call analytics. It runs `get_call_analytics` in one go. You get the total count and average duration in one clean block of data. That's the difference.
What your AI can actually do with this
Marchex MCP Server gives your AI client full control over conversation intelligence. You can analyze call outcomes and track marketing performance by having your agent interact directly with Marchex's data using specific tools.
Searching and Retrieving Call Records
Your agent uses search_calls to pull through entire call histories, letting you narrow down records based on criteria like date ranges or phone numbers. Once you find a potentially relevant record, it fetches all available metadata for that single call via get_call_details. This pair of tools lets your client drill deep into specific conversations and get the full picture of what happened.
Analyzing Campaign Performance
To understand how campaigns are doing, your agent first runs through all active and inactive tracking campaign setups by calling list_campaigns. If you need to know exactly how a particular campaign is configured or if its status changed, it uses get_campaign_details for that specific setup. For the big picture, get_call_analytics calculates and returns aggregated call performance metrics across any given time period, giving you an overall view of campaign health.
Managing Phone Numbers and Accounts
Your client handles everything related to infrastructure management. It starts by listing every phone number that can be used for tracking calls using list_numbers. When you need specific operational details on one of those numbers—like its current status or associated metadata—it pulls that info with get_number_details.
For the user side, your agent can generate a simple listing of all users associated with the platform via list_users, and it also has access to list_accounts to return a list of every user account linked to Marchex. When you need deep insight into an individual organization or user's details, it runs get_account_details.
Putting It All Together
Your agent handles the entire lifecycle flow. You can check campaign setups with get_campaign_details and then use that context to run broad performance reports using get_call_analytics. The system maintains a clear separation between user management (list_users, list_accounts) and number/system management (list_numbers, get_number_details). Your AI client doesn't require you to jump through multiple dashboards; it just takes your natural language request—like, “What were the metrics for Q3?” or “Find me all calls from last Friday using this number”—and executes the precise sequence of tool calls needed to get the data.
It’s direct. It’s immediate.
019d75ce-fe6a-7109-849f-bfd0545d34a2 
Here's how it actually works
The bottom line is that you use natural conversation; the server handles all the API calls and data structuring.
Subscribe to the server and provide your Marchex Client ID and Secret.
Connect your preferred AI client (Claude, Cursor, etc.) to the Vinkius Marketplace.
Ask your agent a question—for example: 'Show me call analytics for last week'—and let it run the necessary tools automatically.
Who is this actually for?
Marketing Ops Managers, Support Engineers, and Data Analysts. You're the person who gets frustrated having to manually cross-reference a campaign ID from one sheet with call records in another dashboard. This server lets you run complex diagnostic queries instantly.
Running reports on which campaigns performed best last quarter, linking specific calls to overall revenue metrics.
Diagnosing why a customer service call failed or tracking the full history of interactions across different numbers.
Checking account health and ensuring that specific lead campaigns are correctly configured before launch.
What Changes When You Connect
Pinpoint failure points fast. Instead of checking dashboards, ask your agent to run search_calls for a date range and immediately narrow it down with specific criteria like an incomplete conversation or duration under 30 seconds.
Understand campaign health in real-time. Use get_call_analytics to see total call counts and average duration without writing complex SQL queries, just by asking your agent.
Diagnose setup issues instantly. If a number isn't working, don't guess; run get_number_details to pull the current status and configuration of that specific tracking line.
Audit who did what. Use list_users and get_account_details when you need to confirm which user or account was responsible for a set of actions or data access.
Compare campaign performance easily. Run list_campaigns, then use get_campaign_details on the suspects, letting your agent compare their settings side-by-side in a single chat window.
See it in action
Investigating a Failed Campaign Launch
The Marketing Manager notices 'Q3 Lead Gen' campaign calls are dropping. They ask the agent to run list_campaigns first, confirming the ID. Then, they instruct it to use get_campaign_details and finally run search_calls for that campaign's period to find the pattern: all failed calls originated from a specific number listed via list_numbers. The root cause is found in minutes.
Auditing User Access and Performance
The Support Lead needs to know if a certain employee (found using get_account_details) was responsible for all calls logged last month. They ask the agent to use list_users to confirm IDs, then run search_calls filtered by that user's timeframe and cross-reference call results with overall metrics from get_call_analytics.
Determining Call Failure Source
A customer reports a bad experience. The agent uses the phone number provided to run get_number_details. They then use that number in search_calls, and finally feed those results into get_call_details to pull up the exact call transcript metadata, solving the issue without needing CRM access.
Comparing Campaign Structures
A junior analyst needs to know why Campaign A performs better than Campaign B. Instead of looking at two dashboards, they ask the agent to run get_campaign_details for both campaigns sequentially and compare key fields like landing page URL or tracking method.
The honest tradeoffs
Assuming a single search works.
The user tries to ask, 'Show me all call metrics.' The agent only runs one tool and gets an incomplete view, leaving the user guessing if they missed something vital in the setup or number data.
You need multiple steps. First, run list_numbers to confirm valid numbers exist. Then, use search_calls with those confirmed numbers, followed by running get_call_analytics on the results.
Ignoring account context.
The user runs a general analytics query without specifying which client or organization they are looking at. The result is vague data that doesn't match their current project scope.
Always start by running list_accounts to ensure you have the correct, scoped account ID before calling any other tools like get_call_analytics.
Bypassing details with a list view.
The user runs search_calls and sees 10 call IDs. They assume the search result summary is enough, but they don't know if those calls were actually marked as 'successful' or 'abandoned'.
After using search_calls, you must run get_call_details for a few sample IDs to confirm all the necessary metadata (like status codes) before making any conclusions.
When It Fits, When It Doesn't
Use this server if your core problem revolves around phone calls, campaign attribution, and call performance metrics. You need to link 'who called' with 'what happened on the call.'
Don't use it if you are trying to manage complex billing cycles or update user passwords; those require separate CRM write-back tools.
If your goal is simply a raw list of all users, list_users works. But if you need to know what that user did (their calls), you must chain the tools: use get_account_details first, and then pass that context into search_calls. If you only care about marketing spend vs. call volume, you'll need to manually feed data from your ad platform alongside the metrics pulled by get_call_analytics; this server doesn't track billing itself.
Questions you might have
How do I use get_call_analytics? +
You tell your agent which time frame you want metrics for. The server runs get_call_analytics and gives you aggregate numbers like total calls or average duration.
What is the difference between list_campaigns and get_campaign_details? +
list_campaigns just spits out a roster of all campaign names. You must use get_campaign_details next, providing the specific ID, to see its actual setup.
Can search_calls find calls from last year? +
Yes, as long as you specify a date range that includes those records. You need to provide enough context (date/number) for search_calls to work.
Do I need get_account_details before searching calls? +
No, but it's good practice. If you run get_account_details, your agent knows which account scope to limit the search to when running search_calls.
How do I handle authentication when running list_accounts? +
You must provide your Marchex Client ID and Secret credentials. The server uses these to authenticate your AI client against the platform, ensuring secure access before listing accounts.
What specific metadata does get_call_details return for a single call? +
It returns detailed records including the caller's number, connection duration, final status (e.g., answered, busy), and which campaign triggered the call.
If I run many searches, how does the server handle rate limiting for search_calls? +
The Vinkius platform manages standard API rate limits. We recommend your agent implements smart retry logic or groups related calls into fewer, larger batches to avoid throttling.
What error message should I expect if the ID used in get_campaign_details is invalid? +
You'll receive a clear API error (likely 404) stating that the specified resource does not exist. This lets your agent recognize bad input and adjust its workflow without crashing.
How do I find my Marchex Client ID and Secret? +
Log in to the Marchex Developer Console and create an application to receive your credentials.
What call data can I access? +
You can access call metadata, duration, caller info, status, and aggregated performance metrics.
Is my integration secure? +
Absolutely. Your credentials are encrypted at rest and injected securely at runtime.
We've already built the connector for Marchex. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 10 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.