Metadata Assistant

Generate titles, alt text, captions, descriptions and tags in bulk, in your language, with a local model or a cloud provider.

Automation

Titles, alt text, captions and tags for a whole library, written for you to correct.

Alt text is the field everyone knows they should fill in and almost nobody fills in, because writing two thousand descriptions by hand is not a task anybody starts. So it stays empty, the library stays unsearchable, and the accessibility audit stays uncomfortable.

The Metadata Assistant writes a first pass for you. Titles, alt text, captions, descriptions and tags, in the language you choose, across as many images as you select. What was an impossible job becomes an afternoon of correcting.

You choose the engine. A local model runs on your own machine with no key and no cost, which is the right default for a big library and for anything involving people. OpenAI and Google Gemini are available if you have a key and want their phrasing.

One honest caveat, and it is the important one: a generated description knows what is in the picture but not why the picture is on the page. It can see a person at a desk; it cannot know that the photograph is there to show the desk. So treat the output as a draft, and correct the images that carry real information. That is still an enormous saving over starting from nothing.

Play Video

Where to find it

Automation, then Images, then Metadata Assistant

What it replaces

Two thousand empty alt attributes and the guilt that goes with them

What you get

Bulk generation of title, alt text, caption, description and tags
Local model, OpenAI or Gemini
Language selection
Prefix and suffix with variables
Selection by search and filter
Part of the free plugin

Requirements

Part of the free plugin. Works with a local model, or with OpenAI or Gemini if you have a key.

Frequently asked questions

As a first pass, yes, and it is vastly better than empty. But context decides what alt text should say, and only you know the context, so correct the images that carry real information. There is a piece on the blog about which images those are.

Only if you choose a cloud engine. The local model runs on your own machine with nothing uploaded, which is the right choice for photographs of real people.

They should have an empty alt attribute, not a description, so a screen reader skips them silently. Generating text for a background texture makes a page worse to listen to, so exclude those from the run.

Features

Five fields at once

Title, alt text, caption, description and tags, each switchable, across a whole selection.

Features

Local or cloud

A local model with no key and no per-image cost, or OpenAI and Gemini if you prefer their output.

Features

Your language

Descriptions written in the language your site is actually in.

Features

Prefix and suffix

Add fixed text around generated values, including variables like the site name, for consistent naming across a library.

Experience Metadata Assistant in WunderPaint

Titles, alt text, captions and tags for a whole library, written for you to correct.

Brush & Paint Styles

Paint with six styles that mix pigment instead of blending it, pick from thirty-seven tips shown as real strokes, draw a tip of your own out of shapes, and let the color travel along the stroke.

Collage & Photo Grid

Arrange media library photos as a grid, mosaic, polaroid, filmstrip or contact sheet, tune gaps, corners and frames on a live preview, and insert it all as an editable layer group.

360° Panorama

Describe a place and the tool builds the full sphere around you, or let it rebuild a photo you already have as the whole scene around the camera. What comes out is somewhere visitors can look around, hotspots and all, in an embed that carries its own viewer.

AI Studio

Generate, edit, retouch and clean up images from a prompt or a single click, all in one panel. Many tools run locally with no key; prompt generation uses your own Gemini or OpenAI.

Asset Generator

Gradients, text treatments, shapes and illustrations generated from a prompt, in your brand colors, arriving as editable layers.

Stock Images

Search Pexels, Pixabay and Unsplash without leaving the editor, and insert what you find as a layer rather than a download.

Asset Library

Ready-made designs across social, marketing and seasonal categories, opening as layered documents in a new tab so your current work stays untouched.

Brand Kits

Colors, typefaces, a logo, a watermark and a company profile in one place, feeding every picker, template and generator on the site.

Media Library Manager

Folders and tags that persist outside the editor, semantic search that finds images by what they show, color and orientation filters, smart folders, and a full set of housekeeping tools.

Check Contrast

Audits the text layers in your document, reports the contrast ratio for each and tells you which ones are hard to read.

QR Code Generator

Links, text, email, phone, SMS, Wi-Fi access and contact cards, styled with your colors, module shapes and a center logo.

Chart & Table

Turn data into chart or table layers that stay editable, styled with your own typeface and colors rather than a spreadsheet default.

Content Generator

Article, SEO title, meta description, slug, tags, FAQ and sources, plus body images, a featured image and a social crop, all as a draft.

Generate from CSV

Type placeholders into your text layers, pick a spreadsheet whose columns carry the same names, and every row renders the design once.