API
An AI agent for your REST API
Give it an OpenAPI document and every operation becomes a tool. No document? Describe the endpoints and they compile the same way.
What you get
REST API capabilities, under your rules
- ✓ API key in a header or query parameter, bearer, basic, or OAuth2 client credentials
- ✓ Static headers sent with every request
- ✓ A health check path to confirm the connection is alive
- ✓ Only public hostnames — private addresses, loopback and cloud metadata endpoints are refused
- ✓ Hostnames re-checked on every request and every redirect, not trusted once
- ✓ Per-tool timeouts and rate limits
What you need to connect it
- Base URL — the root the paths hang off, such as https://api.example.com/v1
- Health check path — defaults to the base URL
- Authentication, and any additional headers the API expects
Discovered tools
What a REST API connection turns into
discovery · REST API
ready- get_orders read
- post_orders create
- delete_orders_id delete Approval
Illustrative. Without a key of your own, a described endpoint is named from its method and path.
Human approval
Destructive work waits for a person
approval
waiting- Tool
- delete_orders_id
- Operation
- DELETE
- Runs against
- Orders API
- Agent
- Operations Agent
Exactly these values
{
"id": "4471"
}
Your decision applies to these values only. If anything about the call changes, it stops for approval again.
The approval covers this call with these arguments. If the model re-plans, it has to ask again.
Connect REST API 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.