Node reference
HTTP Request node: call any API, or give an agent a tool
http_request · Updated September 17, 2026
Instead of hundreds of integrations, Novalink has one HTTP Request node. Use it as a step in the workflow, or connect it to an agent's tools port and let the agent call it.
When to use it
- Fetching data from, or sending results to, any JSON API.
- Giving an agent live data it decides when to fetch.
Ports
- Inputs
in- Outputs
outerror- Tool
attachable to an agent's tools port
Configuration
- Method
method - GET | POST | PUT | PATCH | DELETE · default "GET"
- URL
urlrequired - template
- Headers
headers - object
- JSON body
body - object
- Timeout
timeout_seconds - number · default 30
- Tool description
tool_description - stringShown to the agent when this node is attached as a tool.
- Tool arguments (JSON Schema)
tool_args_schema - objectReference the model's arguments with {{ args.<field> }}.
Output
As a step, fires out with status, ok, headers and body (parsed as JSON when possible). A non-2xx status fails the node, so route it with the error_port error mode if you want to handle it.
As a tool, it is never scheduled. The agent receives the response body, or the status and body when the call failed.
Example
URL: https://api.example.com/orders/{{ args.order_id }}
Tool description: Look up an order by its id.
Tool arguments: { "type": "object", "properties": { "order_id": { "type": "string" } }, "required": ["order_id"] }Good to know
- Private, loopback and link-local addresses are refused, and redirects are not followed.
- Responses are capped in size, and tool responses sent to the model are truncated to keep prompts small.
- See credentials and security.
Keep reading
- Web research agent with an API toolGive a Claude agent an HTTP tool it can call on its own: a research workflow that fetches data from an API and returns a sourced summary.
- Agent node: Claude with tools and structured outputThe Agent node calls Claude with your prompt, can call attached HTTP tools in a loop, and returns text plus schema-validated structured output for branching.
- Execution controls: retries, timeouts and error handlingControl how Novalink runs AI workflows: per-node retries with backoff, timeouts and error modes, and run-wide parallelism, timeouts and token caps.
Build it on the canvas
Create a free account and run your first workflow in the browser, or run Novalink on your own machine.