Use {baseDir}/index.js to call RapidAPI with templates from {baseDir}/templates.
Prefer this skill when the task involves RapidAPI endpoints or template-defined actions.
What This Skill Actually Does
This skill is a minimal RapidAPI client that turns RapidAPI endpoint definitions into callable actions. It is meant for:
- converting a RapidAPI endpoint into a stable action name
- standardizing inputs into query/body/header/path params
- returning a consistent
ok/status/data/error/metashape
It is not a server. It is a small local client you can call from scripts or other skills.
Key Capabilities
- Auto-registers templates from
{baseDir}/templates/*.json listActions()enumerates all registered actions with schemascallAction(name, params)calls a template-defined endpointcallRapidApi(payload)allows direct RapidAPI calls without a templatescripts/import-endpoint.jsconverts a RapidAPI endpoint JSON payload into a template file
Basic Usage
Use config-driven init (recommended):
import { createRapidApiSkill } from "{baseDir}/index.js";
import config from "{baseDir}/config.json" assert { type: "json" };
const skill = await createRapidApiSkill({ config });
const res = await skill.callAction("get_user_tweets", {
user: "2455740283",
count: 20
});
Or direct call (no template):
const skill = await createRapidApiSkill({ config });
const res = await skill.callRapidApi({
host: "twitter241.p.rapidapi.com",
path: "/user-tweets",
method: "GET",
query: { user: "2455740283", count: 20 }
});
Template Design Notes
Templates are plain JSON. They should contain:
name,host,path,methodquerySchema(and optionallybodySchema,headerSchema,pathParams)
Example snippet:
{
"name": "get_user_tweets",
"host": "twitter241.p.rapidapi.com",
"path": "/user-tweets",
"method": "GET",
"querySchema": {
"user": {"type": "string", "required": true},
"count": {"type": "number", "required": true},
"cursor": {"type": "string"}
}
}
Where It Fits
Use this skill when you need a consistent, reusable RapidAPI interface without building a backend. It is especially useful for:
- social data APIs (X/Twitter, TikTok, LinkedIn)
- search/aggregation APIs
- repeated RapidAPI calls across multiple tasks or workflows