FRENES

Glossary · MCP protocol

Webhook

A webhook is a mechanism that lets a service automatically send a notification to a web address when an event occurs. It can tell your AI agent that someone submitted a form, for example, instead of requiring the agent’s application to keep checking for new submissions.

Webhook: the explanation

Setting up a webhook usually involves selecting events and providing an address that can receive requests. When a selected event happens, the sending service delivers data to that address. The receiving application checks the request and processes it or schedules further work. The payload may include useful details or simply identify a resource to retrieve. This event notification pattern is described in the documentation: https://docs.github.com/en/webhooks/about-webhooks. It reduces some repeated checks without eliminating all communication between the services involved.

Webhooks connect events to subsequent steps in an automated workflow. A new upload might trigger document classification, while a completed transcription could start a summary. The recipient does not need to be the AI model itself. An intermediate application can inspect the event, remove unnecessary information, and apply rules before involving an agent. Keeping those stages separate makes it easier to control when AI is used and which actions are allowed to follow from an external notification.

Receiving a webhook does not by itself establish that a message is authentic or that its contents authorize an action. The integration needs to verify the sender using the service’s supported mechanism, often a signature. It also needs to handle failures, delays, and repeated deliveries. Ordering and retry behavior vary between services. An important operation should therefore remain safe and consistent when an event arrives late, arrives twice, or requires additional checking before it is processed.

A concrete example

You collect support requests through an online form. Each submission triggers a webhook to your case management application. The application verifies the notification, passes the relevant text to an agent, and prepares a draft reply for your review. If the service later delivers the same notification again, the application recognizes the existing case. It avoids opening a duplicate request or sending the customer another response to an issue already handled.

Why it matters

Webhooks can make an agent more responsive, but they also introduce information from external systems. Check which events trigger activity, which data is included, and how repeated deliveries are handled. For sensitive operations, separate notification from permission to act. Learning that a document has arrived does not automatically authorize the agent to share it, delete it, or obey instructions embedded in its contents or accompanying metadata.

Common misconceptions

  • A webhook does not guarantee immediate delivery exactly once: network conditions and service behavior can cause delays, repeated notifications, or retries before the receiving application can successfully process the event and complete any associated work.
  • A webhook payload is not automatically an instruction the model should obey: its contents may come from external users and must be handled according to the application’s permissions, validation rules, and intended workflow for that event.

Related terms: mcp protocol

MCP (Model Context Protocol)
MCP, or Model Context Protocol, is a protocol for exchanges between an AI application and services that provide data or functions.
MCP server
An MCP server is a software component that exposes data or functions through the Model Context Protocol.
MCP client
An MCP client is the component of an application that communicates with a server using the Model Context Protocol.
API
An API, or application programming interface, defines how software can use another component’s functions or data.
SDK
SDK stands for software development kit.
structured output
Structured output is a response arranged according to a defined format, with expected fields and data types.
A2A protocol
The A2A protocol is an open set of rules that lets independent AI agents communicate and delegate tasks.

Where this site covers it

The term appears on 28 pages of this site. A few of them:

Go further

Frequently asked questions

How is it different from an API?

An API defines an interface between programs. A webhook notifies another program about an event by sending a request to a designated address. They often work together, with the notification prompting an additional API request.

Do I need my own server?

You need a receiving endpoint that the sending service can reach. Your application provider, hosting platform, or automation service may supply it. You do not necessarily need to operate and maintain a separate server yourself.

How do I prevent duplicate actions?

The receiving application can record event identifiers or check the current state before acting. The right approach depends on the service. The aim is to prevent repeated notifications from repeating an operation with practical consequences.

Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score