Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/nayalsaurav/ai360/llms.txt

Use this file to discover all available pages before exploring further.

AI360’s API is implemented as Next.js App Router API routes. Every endpoint lives under /api/utilities/ and acts as a thin, secure proxy between the Next.js frontend and upstream AI services — Google Gemini 2.5 Flash, Cloudflare Workers AI, and ImageKit. Because all service credentials are stored as server-side environment variables, client code never touches an API key. You call the route, the route calls the AI provider, and the result comes back in a consistent JSON shape.

Base URL

EnvironmentBase URL
Developmenthttp://localhost:3000
ProductionYour deployment URL (Vercel or custom domain)
All routes use the POST method unless explicitly noted. Paths are relative to the base URL.

Endpoints at a glance

POST /api/utilities/summarizer

Summarize any block of text into a concise, meaningful paragraph using Gemini 2.5 Flash.

POST /api/utilities/cold-email

Generate a short, persuasive cold email from a context prompt using Gemini 2.5 Flash.

POST /api/utilities/email-template-generator

Produce a complete HTML email template from a prompt, ready to drop into any email client.

POST /api/utilities/image-generator

Generate an image from a text prompt via Cloudflare Workers AI (FLUX.1 Schnell). Returns a base64-encoded PNG data URL.

POST /api/utilities/background-remover

Remove the background from an uploaded image using ImageKit’s processing pipeline.

POST /api/utilities/code-explainer

Explain a code snippet in plain language using Gemini 2.5 Flash.

POST /api/utilities/flowchart-generator

Generate a structured flowchart JSON from a natural-language description.

POST /api/utilities/resume-analyzer

Analyze a PDF resume against a job description, returning ATS, tone, content, structure, and skills scores. Accepts multipart/form-data.

GET /api/upload-auth

Return time-limited ImageKit upload credentials (token, expire, signature) for client-side direct uploads.

Authentication

All routes are server-side only. Upstream service credentials are injected at build/runtime via environment variables and are never exposed to the client:
VariableUsed by
GEMINI_API_KEYAll Gemini-powered routes
CLOUDFLARE_IDImage Generator (Cloudflare Workers AI)
CLOUDFLARE_API_KEYImage Generator (Cloudflare Workers AI)
IMAGEKIT_PRIVATE_KEYBackground Remover, Upload Auth
Client requests do not need to pass any Authorization header or token — the server handles authentication with each upstream service transparently. For environment setup instructions, see the Configuration page.

Content types

RouteRequest content type
Most endpointsapplication/json
/api/utilities/resume-analyzermultipart/form-data (FormData)
/api/upload-authNo request body required
For multipart/form-data requests, do not set the Content-Type header manually. Let the browser or fetch set it automatically so the MIME boundary is generated correctly.

Error format

All routes return errors as a JSON object with a single error key containing a human-readable message:
{
  "error": "Human-readable error message"
}
Standard HTTP status codes apply:
StatusMeaning
400Missing or invalid request input (e.g., empty required field, wrong file type)
500Upstream service failure (Gemini, Cloudflare, or ImageKit returned an error)
None of the API routes include rate limiting in the open-source codebase. If you deploy AI360 publicly, every endpoint is open to unbounded usage which will rapidly exhaust your Gemini, Cloudflare, and ImageKit quotas. Add a rate-limiting layer — such as Vercel’s Edge Middleware, Upstash Ratelimit, or an API gateway — before exposing these routes to the internet.

Next steps

Resume Analyzer

Deep dive into the most advanced endpoint — structured JSON output, PDF ingestion, and multi-category ATS scoring.

Configuration

Set up your environment variables for Gemini, Cloudflare, and ImageKit so every route is ready to use.

Build docs developers (and LLMs) love