Quickstart

Get up and running with Picora — the resource hosting platform for AI workflows — in under 5 minutes.

This page covers the universal happy path: create an account, get an API Key, upload your first image. For deeper task-oriented guides, see the User Guide; for AI assistant integration, see AI Workflows.

1. Create an account {#step1}

Pick the platform that matches your region:

Platform	URL	Default language	Compliance
Overseas	center.picora.me/register	English	GDPR
Mainland China	center.picora.cn/register	中文	PIPL (个人信息保护法)

Sign up with email + 6-digit OTP (fastest), or password, or Firebase Google/Apple (overseas), or WeChat / SMS (mainland). The trial plan auto-applies — 100 MB image storage, no credit card.

For the full breakdown of registration methods, see Account → Registration.

2. Pick what to upload {#step2}

Picora hosts four resource types. Trial users get images only; pro / pro_plus unlock the rest:

Type	Trial	Pro / Pro+	Guide
Images	✅ 100 MB	✅ 5 GB / 50 GB	Images
Videos	—	✅ HLS adaptive streaming	Videos
Audio	—	✅ ID3 cover extraction	Audio
Documents (Markdown / Typst)	—	✅ Markdown auto image rewriting; Typst rendered in-browser	Documents

This quickstart focuses on images; replace /v1/images with /v1/videos / /v1/audio / /v1/docs for other types.

3. Get your API Key {#step3}

For programmatic uploads (scripts / tool clients like PicGo / Moraya):

1. Open Settings → API Keys
2. Click Create new key
3. Name it (e.g., “blog uploader”) and pick scopes (default: read + write, no delete)
4. Copy the displayed sk_live_... key now — it’s shown once

Caution

The API Key is shown only at creation. If lost, revoke and create a new one. See API Keys management for security best practices.

For AI assistants (Claude Desktop / Cursor / Moraya), use OAuth instead — it’s zero-install and easier to revoke. Skip to step 5.

4. Upload your first image {#step4}

Recommended — the official @picora/sdk (typed, auto-retries rate limits, covers the whole API):

import { createPicoraClient } from '@picora/sdk'   // npm install @picora/sdk
import { readFileSync } from 'node:fs'

const picora = createPicoraClient({ apiKey: process.env.PICORA_API_KEY })

const img = await picora.images.upload({
  file: readFileSync('photo.jpg'),
  filename: 'photo.jpg',
  isPublic: true,
})
console.log(img.url)      // https://media.picora.me/<nanoid>.webp — your permanent CDN link

Or call the REST API directly with curl (no install needed):

curl -X POST https://api.picora.me/v1/images \
  -H "Authorization: Bearer sk_live_YOUR_KEY" \
  -F "file=@./photo.jpg" \
  -F "isPublic=true"

Successful response (curl):

{
  "success": true,
  "data": {
    "id": "xK9mR2pQ7vB",
    "url": "https://media.picora.me/xK9mR2pQ7vB.webp",
    "filename": "photo.jpg",
    "sizeBytes": 102400,
    "isPublic": true,
    "createdAt": "2026-04-27T08:00:00.000Z"
  }
}

The url is your permanent CDN link. Share it anywhere; it never changes.

For mainland China users, replace api.picora.me with api.picora.cn and media.picora.me with media.picora.cn.

5. Connect your tool / AI assistant {#step5}

Pick the path that matches your workflow:

For Markdown editors

Editor	Method	Guide
PicGo	Custom Web Uploader	PicGo
Moraya	JSON config paste / one-click deep link	Moraya
Typora	Custom command	Typora
Obsidian	Image Auto Upload Plugin	Obsidian

For AI assistants (Claude / Cursor / Moraya AI)

Mode	Setup	Guide
stdio (local)	Run @picora/mcp-server with an API Key	MCP → Quickstart
Remote (zero-install)	Add https://mcp.picora.me to your client and authorize in the browser	MCP → Remote

Both paths are coming soon — the npm package is unpublished and the remote endpoint is not yet deployed. See MCP for status.

After connecting, your AI assistant can call Picora tools like:

“Upload the README I’m editing to Picora and give me a public link.”

“List my images uploaded this week, sorted by size.”

See Prompt templates for 10 battle-tested examples.

6. Done! What’s next?

Your images are now hosted on Picora with permanent, clean URLs. Each image gets a unique 11-character nanoid (e.g., https://media.picora.me/xK9mR2pQ7vB.webp).

Where to go from here:

• User Guide — comprehensive task-oriented manual (account / billing / quota / advanced)
• AI Workflows — full reference for MCP integration with Claude / Cursor / Moraya
• TypeScript SDK — @picora/sdk, the recommended way to build integrations
• API Reference — REST API spec for all endpoints
• API Explorer — interactive Scalar try-it-out
• Plan comparison — when to upgrade
• FAQ — common questions