Developers
Webhooks when an agent finishes, fails, or needs a decision
A webhook posts to a URL you own when something happens in your workspace. Every delivery is signed with an HMAC of the timestamp and the body, using a signing secret shown once when you create the endpoint.
What you get
Webhooks capabilities, under your rules
- ✓ Every delivery signed with an HMAC of the timestamp and the body
- ✓ A timestamp on every delivery, so a captured request cannot be replayed later
- ✓ Failed deliveries retried with a backoff
- ✓ Every event carries an id, so repeats can be ignored
What you need to connect it
- An HTTPS endpoint you own that answers 2xx quickly
- Somewhere to keep the signing secret, which is shown once
Human approval
Destructive work waits for a person
approval
waiting
delete_customers
CRITICAL risk
Destructive
- Tool
- delete_customers
- Operation
- DELETE
- Runs against
- Production PostgreSQL
- Agent
- Support Agent
Exactly these values
{
"filters": {
"id": 100
}
}
Your decision applies to these values only. If anything about the call changes, it stops for approval again.
Connect Webhooks to your first agent
Connect one system and read the agent Caterfli writes from it. There is a demo connection too, if you would rather look before you wire anything up.