No-code · Flowise
Flowise + SerpKite
Build a SerpKite Custom Tool in Flowise with an input schema and a few lines of JavaScript, then connect it to a Tool Agent.
Flowise is no longer developed
FlowiseAI archived the Flowise repository in August 2026, so it takes no new nodes and there won't be an official SerpKite node. The Custom Tool below keeps working on existing Flowise installs. For a maintained visual builder, see the SerpKite node for n8n or the Dify plugin.
Overview
What Flowise is, and where SerpKite fits
Flowise is an open-source visual builder for LLM flows and agents. A Custom Tool is a JavaScript function with a declared input schema; the agent fills the inputs, and the function's return string goes back to the model.
In the function, inputs are available as $<property> variables (here $query), and Flowise variables as $vars.<name>, so the API key stays out of the flow.
Setup
Set it up in 5 steps
-
1
Get an API key
Sign up (no card), create a key in the dashboard and export it as
SERPKITE_API_KEY. New accounts get 2,500 free credits, then 1,000 every month.export SERPKITE_API_KEY=skt_live_… -
2
Store the key as a variable
In Flowise open Variables → Add Variable. Name:
serpkiteApiKey, type Static, value: your key. -
3
Create the Custom Tool
Go to Tools → Create. Name
google_search, a clear description, and one input property:query(string, required). -
4
Paste the function
Use the JavaScript below as the tool function and save.
-
5
Connect it to an agent
In a chatflow, add a Custom Tool node, select
google_searchand connect it to a Tool Agent with a chat model.
Code
Custom Tool function
// Flowise Custom Tool. Input schema: query (string, required)
const res = await fetch("https://api.serpkite.com/v1/search", {
method: "POST",
headers: {
Authorization: `Bearer ${$vars.serpkiteApiKey}`,
"Content-Type": "application/json",
},
body: JSON.stringify({ q: $query, country: "us", format: "markdown" }),
});
if (!res.ok) {
const { error } = await res.json();
return `Search failed: ${error.message}`;
}
return await res.text(); // Markdown: results, People Also Ask, related searchesAI assistants
Ask ChatGPT, Gemini and other AI assistants
Every assistant is one POST: https://api.serpkite.com/v1/chatgpt (or /v1/gemini, /v1/copilot, /v1/ai_mode, /v1/brave_ai) with the same Bearer header and a JSON body with prompt. Create a second Custom Tool (ask_chatgpt, input prompt) with the function below. The response has the answer (text, markdown), cited sources, products and, with track, mentions of your brands and domains. See AI assistants.
Through the MCP server the same assistants are the chatgpt, gemini, copilot, ai_mode and brave_ai tools.
ChatGPT 4, Gemini 3, Copilot Search 3, Google AI Mode 2 and Brave AI 2 credits per answer; failed, refused and empty answers are free. Brave answers in 1–10 s, ChatGPT in 10–30 s and Gemini in 5–30 s, so allow at least 120 seconds per call.
// Flowise Custom Tool. Input schema: prompt (string, required)
const res = await fetch("https://api.serpkite.com/v1/chatgpt", {
method: "POST",
headers: {
Authorization: `Bearer ${$vars.serpkiteApiKey}`,
"Content-Type": "application/json",
},
body: JSON.stringify({ prompt: $prompt, format: "markdown" }),
});
if (!res.ok) {
const { error } = await res.json();
return `ChatGPT failed: ${error.message}`;
}
return await res.text(); // the answer as Markdown, with its sourcesFAQ
Flowise and SerpKite: common questions
Keep exploring
Related APIs and integrations
Start building
Give your Flowise project Google search
2,500 free credits, then 1,000 every month. No credit card.