Convoy MCP for AI. Orchestrate webhooks with natural language prompts.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Convoy manages your entire webhook infrastructure from within your AI client. Create, update, and delete endpoints; broadcast events to all subscribers; or pause traffic instantly for maintenance.
It gives you programmatic control over complex event delivery pipelines without ever leaving your terminal.
What AI agents can do with Convoy Automation
Activate endpoint
Restores traffic flow to a specific webhook destination that was paused.
Batch retry event deliveries
Attempts to resend multiple failed events in one batch process.
Broadcast event
Sends an event payload to all matching endpoints simultaneously.
Create, read, update, or delete specific webhook destinations within a project.
Trigger events across multiple services simultaneously or send a single event to all subscribed endpoints.
Instantly pause specific webhooks or activate them when maintenance is complete, managing live traffic flow.
Retrieve detailed logs of past delivery attempts to debug failure points in your pipeline.
Process large numbers of endpoints, subscriptions, or filters at once using bulk tools.
Ask an AI about this
Waiting for input…
What AI agents can do with Convoy: 27 Tools for Webhook Management
Use these specific tools to perform every action related to webhook infrastructure, from creating endpoints to debugging complex event delivery failures.
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 Convoy on VinkiusActivate Endpoint
Restores traffic flow to a specific webhook destination that was paused.
Batch Retry Event Deliveries
Attempts to resend multiple failed events in one batch process.
Broadcast Event
Sends an event payload to all matching endpoints simultaneously.
Bulk Create Filters
Creates multiple data filters for a subscription using a single command.
Bulk Onboard
Adds many new webhook endpoints and subscriptions to your project at once.
Create Endpoint
Registers a brand new destination URL for incoming webhooks.
Create Event
Manually generates and sends a single, specific event payload.
Create Event Type
Defines the schema or structure for a new type of event that your system uses.
Create Filter
Sets up rules to determine which events should trigger specific subscriptions.
Create Portal Link
Generates a dedicated URL for accessing webhook management settings.
Create Source
Defines the origin or system that is responsible for sending events.
Create Subscription
Sets up a connection between an event type and a destination endpoint.
Delete Endpoint
Permanently removes a webhook destination from the project.
Dynamic Event
Creates a flexible event whose content can be determined at runtime.
Fanout Event
Disperses an event to multiple, unrelated endpoints efficiently.
Get Delivery Attempt
Retrieves the detailed status of one specific attempt to deliver an event.
Get Endpoint
Fetches all metadata about a single, existing webhook destination.
Import Event Types
Loads defined event types into the system using an OpenAPI specification file.
List Delivery Attempts
Retrieves a list of all delivery attempts associated with one specific event delivery.
List Endpoints
Lists every webhook destination configured in the current project.
List Event Deliveries
Shows a summary list of all events that have been delivered recently.
List Meta Events
Retrieves general system-level events, separate from application data.
List Subscriptions
Lists all active subscriptions that link event types to endpoints.
Pause Endpoint
Temporarily stops traffic to a specific webhook endpoint immediately.
Retry Event Delivery
Re-sends an entire recorded event delivery that failed previously.
Update Endpoint
Changes the URL or settings of an existing webhook destination without deleting it.
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 Convoy 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 Convoy, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ 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 Convoy. 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
Built on the Model Context Protocol (MCP) for 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 26 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Webhook debugging is usually a nightmare of dashboards and copy-pasting IDs., Solved with Vinkius AI Gateway
Right now, when an integration fails, you're stuck. You jump between the webhook dashboard, the logging service, and your IDE. You have to manually find the endpoint ID, check its status, then drill down into a specific delivery attempt just to see if the data even got there. It’s tedious, slow, and prone to human error.
With this MCP, you simply tell your AI client what went wrong. Your agent handles all the necessary checks, like listing endpoint statuses or retrieving failure reports. You get an immediate, actionable diagnosis without ever leaving your terminal.
Control event delivery with Convoy's full suite of tools.
You no longer have to manually list endpoints and then decide which ones need maintenance. You can tell the system to check all critical paths using `list_endpoints` and then execute a status change with `pause_endpoint`. It manages the operational state for you.
The result is that your entire event pipeline runs under programmatic control. The complexity of managing hundreds of subscriptions becomes as easy as asking a question.
What your AI can actually do with this
Webhook management is usually a headache of dashboards and manual clicking. Convoy changes that by letting your AI client orchestrate every aspect of your event infrastructure through simple conversation. You can build new endpoints, or maybe you need to test an old one, so you use create_endpoint for setup. When the time comes to push data, instead of manually triggering a flow, you just tell your agent to broadcast an event.
The system then handles sending that message to every matching destination. If things go wrong, you don't guess; you check delivery attempts and pinpoint exactly where the failure happened. All this control is available through Vinkius, giving you one place to manage thousands of connections across any MCP-compatible client.
019ea5e6-0197-71c9-a756-70d1a86df4ad 
Here's how it actually works
The bottom line is that you manage complex event delivery infrastructure using natural language prompts instead of clicking through multiple web dashboards.
Subscribe to this MCP and provide your Convoy API Key and Base URL.
Tell your AI client what you need: 'List all endpoints for the production project' or 'Create a new endpoint at X address'.
Your agent executes the necessary commands, confirming success or detailing any failures in real time.
Who is this actually for?
This MCP is for the DevOps Engineer who hates manually checking webhook logs at 3 AM. It's built for Backend Developers who need to test event triggers without leaving their IDE, and SREs needing immediate control over production traffic flow.
Uses the MCP to toggle endpoint states (pause_endpoint / activate_endpoint) and run bulk operations like onboarding new services.
Triggers test events or uses broadcast_event to simulate user actions, allowing them to confirm webhook functionality without deploying code changes.
Checks delivery logs using tools like list_delivery_attempts to diagnose why a critical service failed to receive an expected event.
What Changes When You Connect
Instant traffic control: You don't have to jump into a dashboard just to hit pause. Use pause_endpoint or activate_endpoint to instantly manage live webhook flow for maintenance windows.
Deep debugging visibility: When an event fails, you don't get a vague error code. You can use get_delivery_attempt and list_delivery_attempts to pinpoint the exact moment and reason for failure.
Bulk efficiency: Managing fifty endpoints one by one is slow. Use bulk_onboard or bulk_create_filters to handle massive infrastructure changes in a single conversation.
Complete event lifecycle control: Need to test an old flow? Instead of recreating it, you can use create_event or fanout_event to programmatically trigger and verify the entire sequence.
Simplified setup: Setting up new integrations requires defining types and connections. Use create_event_type followed by create_subscription to define your system's event contracts easily.
See it in action
Testing a high-traffic endpoint rollout
A developer needs to verify that the new payment webhook only receives data after it hits staging. They ask their agent to create_endpoint for the staging URL, and then use broadcast_event with test data. This confirms the full path works before flipping the switch.
Handling a production outage
The SRE notices delivery failures on critical endpoints. They ask their agent to check the status using list_delivery_attempts, find the failing batch, and immediately run retry_event_delivery without touching any UI.
Scaling a new microservice
A team is launching 50 related services that all need to subscribe to user creation events. Instead of running 50 manual steps, they use bulk_onboard and define the necessary subscriptions in one go.
Maintenance mode enforcement
A developer needs to update an endpoint's backend logic. They ask their agent to run pause_endpoint on the primary production URL, ensuring no data is lost or corrupted during deployment, and then reactivate it when done.
The honest tradeoffs
Trying to manage webhooks with simple messaging
Attempting to manually copy/paste endpoint IDs into a general chat window or using basic API calls without context.
Use the dedicated list_endpoints tool first to get all necessary IDs. Then, use specific tools like update_endpoint or delete_endpoint for precise control.
Over-relying on UI dashboards
Logging into the webhook gateway dashboard every time a test needs running just to verify connectivity.
Run automated tests using your agent. Start by calling create_event and then check results with get_delivery_attempt—all from your client.
Ignoring bulk operations
Adding new integrations one endpoint at a time when the requirement is to onboard an entire suite of services.
Use the dedicated tools: bulk_onboard or bulk_create_filters. These save hours of clicking and typing.
When It Fits, When It Doesn't
You need this MCP if your process involves high-volume, critical event delivery where manual intervention is risky or slow. Specifically, use it when you must monitor failed deliveries (using list_delivery_attempts) or control traffic flow instantly (pause_endpoint). Don't use it if you just need to send a simple message; general messaging tools are fine then. However, do not use this for initial schema design; first define your event contracts using create_event_type and import them via OpenAPI specs before setting up subscriptions.
Questions you might have
How do I check if an endpoint is working with Convoy? +
You can use get_endpoint to retrieve the full metadata and status of any specific webhook destination. This confirms its current state before you attempt a test event.
Can I trigger multiple events at once using Convoy? +
Yes, you have two options: use broadcast_event to send the same payload to all matching endpoints, or use fanout_event if you need to dispatch different event types across various services.
What is the difference between create_event and broadcast_event in Convoy? +
Using create_event manually generates a single, specific data payload for testing. broadcast_event, however, sends that event to all configured subscribers simultaneously.
How do I fix an endpoint that failed last week with Convoy? +
You first list the failures using list_delivery_attempts to find the specific delivery ID. Then, you use retry_event_delivery to re-send the entire event batch.
Does Convoy handle endpoint updates? +
Yes. Instead of deleting and recreating a destination URL, you use update_endpoint. This modifies settings while maintaining the existing webhook ID and subscription history.
We've already built the connector for Convoy. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 26 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.