Webhooks for AI agents
Outbound webhooks your agent can fire by name, and inbound URLs that deliver a callback straight back into the running session. No polling loop.
Copy the setup prompt.
Your agent does the rest.
Fetch and execute the appropriate instructions to set me up for superagnt_ from https://mcp.superagnt.com/agent-setup/prompt.md» On by default. The agnt_webhooks_* tools are part of the base tool surface, so a fresh MCP connection already has them.
Everything your agent can do with it
agnt_webhooks_send posts a JSON payload to a workspace endpoint by id or name, so the agent never holds the URL or its secret.
agnt_webhooks_inbound_url returns an ingest URL. Append the session id, hand it to an external service, and whatever that service posts back arrives as a new turn in the same session.
A long third-party job becomes call me back here, instead of a loop that burns tokens waiting for something to finish.
agnt_webhooks_receive_recent lists recent inbound deliveries for the workspace, optionally scoped to one endpoint.
Inbound endpoints verify a per-endpoint signature, so a public URL is not an open door.
tool names03as your agent sees them
agnt_webhooks_sendagnt_webhooks_inbound_urlagnt_webhooks_receive_recentThree jobs it can start on today
Hand off a long job and get woken up
The agent kicks off a render, an export or a human approval elsewhere, gives that service its inbound session URL, and picks up exactly where it left off when the result posts back.
Push results into your stack
When a run finishes, the agent posts the summary to your own endpoint by name, so downstream systems react without anyone watching a dashboard.
Trigger from an external event
A form submission or a CRM change posts to an inbound endpoint and starts the work, instead of a schedule polling for something that may not have happened.
What comes back
Your agent reads this response directly, with no parsing layer of your own in between.
https://mcp.superagnt.com/mcptool callagnt_webhooks_inbound_url{}{
"endpoints": [
{
"id": "whi_3d81",
"name": "render-callbacks",
"ingest_url": "https://api.agntdata.dev/webhooks/ingest/whi_3d81",
"session_url_template": "https://api.agntdata.dev/webhooks/ingest/whi_3d81/{session_id}"
}
]
}What it costs to run
Platform tools carry no per-tool price. What you pay is the AI and credit usage the work causes, on the same workspace balance every other tool draws from.
Each run is itemized, so you can see which agent spent what and on which call.
how credits work →Before you wire it up
01How does an inbound callback reach the right session?
agnt_webhooks_inbound_url returns an ingest URL and a session URL template. Append the current session id, hand that URL out, and the payload posted to it is injected into that session as a new turn.
02Where do endpoint URLs and secrets live?
In the workspace. The agent sends by endpoint name or id, so a URL or a secret never has to appear in a prompt or a tool argument.
03Can I see what came in?
Yes. agnt_webhooks_receive_recent returns recent inbound deliveries for the workspace, filterable to a single endpoint.
04Do I need to enable it?
No. Webhook tools are in the default tool surface for a workspace MCP connection at https://mcp.superagnt.com/mcp.
Point your client at https://mcp.superagnt.com/mcp and your agent has this, plus everything else on one balance.