🎨 MicaCover
Social cover image maker: 14 platform sizes, 8 built-in gradients, layered layout and AI title expansion, backgrounds from the Unsplash / Pexels stock library, export PNG / JPG / WebP.
🎨 Cover settings
Built-in gradients are generated live on the Canvas, with a light noise texture.
Local images stay in your browser and are never uploaded; the config does not keep them, so re-upload after switching devices. Stock photos are fetched by URL and only the URL is saved in the config, so they come back when you reopen it.
Advanced: fit and position
Drag text or the Logo on the canvas to move it; click a layer row or an element on the canvas to select it; double-click the canvas to enlarge it.
Click a layer in the list on the right or text / Logo on the canvas to select it and adjust it here.
Advanced: text details
Text can also be dragged directly on the canvas; use the X / Y number boxes for precise alignment.
Or drop a file here · PNG / JPG / WebP ≤ 20MB
The Logo can also be dragged directly on the canvas; use the X / Y number boxes for precise alignment.
Loading…
Loading canvas…
🧱 Layers
6 layersHow to use MicaCover
Follow the steps to get it done
Write the topic, pick a size and a background
Open the page and start — no registration or sign-in needed.
Click text on the canvas to edit it directly
Settings can be changed at any time; all computation happens in your browser.
Export, save & share the cover from the right panel
Copy or download the result when you are done. Nothing is kept on this site.
Page Actions
Size dropdown, grouped into Chinese platforms and international social media, 14 sizes in total: square 1:1, Xiaohongshu 3:4, Juejin / Zhihu 4:3, X card 16:9, WeChat header, Facebook post / cover, Instagram portrait / story, X header, LinkedIn post / cover, YouTube thumbnail, Pinterest. Switching sizes scales the fonts with the canvas width. Xiaohongshu 3:4 also shows top and bottom margin warning bands in the safe zone.
Switch the background type to a built-in gradient and show the gradient dropdown.
Switch the background type to a solid colour and show the colour picker.
Switch the background type to an image and show the local upload row (list-style) and the 「Get a stock photo」 button, plus the fit controls.
Background gradient dropdown: Ocean, Sunset, Forest, Ink, Dusk, Emerald, Rose, Lavender. Pick one to change the background gradient. Gradients are generated live on the Canvas with a light noise texture.
The colour picker for a solid background. Picking a colour applies it immediately.
The background upload is a file input row: the file name shows on the left (blank when nothing is chosen) and a green "Choose..." button on the right opens the file picker; you can also drag an image onto the row. PNG, JPG, and WebP are supported, with a single image no larger than 20MB. After choosing, the file name shows in the field; click the ✕ next to it to remove the background, or click "Choose..." again to replace it.
The 「Get a stock photo」 button in the image background area. Sign-in required: while logged out the button is greyed out, and clicking it opens the login dialog first — after signing in, the photo picker opens. When it opens, the AI analyses your current theme (main title + subtitle + badge) and suggests the best English search keywords; the top word fills the search box and searches once, and the alternatives sit under the box. Each source returns 10 photos at a time, so 「All」 shows 20.
The keyword input at the top of the photo picker. It starts with the top word from the AI analysis (falling back to a query built from your main title when the AI fails or the theme is empty) and can be edited; press Enter or click 「Search」 to search again with the new keyword.
When the photo picker opens, the AI analyses the theme once (1 point per analysis, free with BYOK; the same theme is only analysed once per page session, so reopening the picker reuses the last result and does not charge again; the per-run cost stays visible under the status line and is hidden for BYOK users). If the AI fails, the picker searches by the main title instead, the cost line shows the failure, and a 「Retry」 button appears — click it to analyse once more.
Suggested keywords under the search box: the AI alternatives (3–4 English words) when the analysis succeeds, falling back to one word from the main title and one from the subtitle when it fails or the theme is empty. Click one to search again with that word; a word identical to the current query is not shown.
The All / Unsplash / Pexels tabs in the photo picker. Each tab shows its own result count (for example 「All (20)」), and the status line shows the split, for example 「Unsplash 10 · Pexels 10, 20 photos in total」. A source without a configured key shows 「Not configured」 and cannot be clicked; it works as soon as the key is set.
The 「Load more」 button under a single-source tab. It adds 10 more photos from that source. It does not appear under the 「All」 tab.
After picking a photo in the picker, click 「Use this photo」 to fetch the full image by URL and set it as the background. If the fetch fails, the status line says so — pick another one.
The 「Remove background image」 button in the image background area. It removes the current background image and stays on the image option, while the canvas temporarily draws the gradient; upload a new image or pick one from the stock library. It only appears while a background image is set.
Cover (fill the canvas, cropping the excess) or Contain (show the whole image, leaving dark bars). Collapsed inside "Advanced: fit and position" under the image upload area.
Horizontal position slider with a number box, 0 to 100, default 50 (centred). Collapsed inside "Advanced: fit and position".
Vertical position slider with a number box, 0 to 100, default 50 (centred). Collapsed inside "Advanced: fit and position".
Background blur slider with a number box, 0 to 20 px, default 0. Blur the background to make the text stand out. Collapsed inside "Advanced: fit and position".
Background dim slider with a number box, 0 to 80%, default 0. Adds a black translucent overlay so text is easier to read. Collapsed inside "Advanced: fit and position".
Dropdown: choose 「Do not enable」 to turn the overlay off; choosing any gradient turns it on with that gradient to darken the background and set off the text. The same 8 presets as the background, default Ink.
Overlay opacity slider with a number box, 0 to 80%, default 35%.
With a text layer selected, edit the main title, subtitle, or badge text here. Multi-line is supported and long text wraps automatically.
Font dropdown for the selected text layer: default sans-serif, serif, or monospace system font stacks.
Font weight dropdown for the selected text layer: Regular 400, Semi-bold 600, Bold 700, Extra-bold 800, Black 900.
Font size number box, 12 to 400 px. Type a value directly; out-of-range values are clamped automatically. Defaults: main title 96, subtitle 40, badge 30 (scaled to the canvas width).
Colour picker for the selected text layer. With 「Auto」 on, the background brightness decides the colour; picking a colour manually turns auto off.
Toggle (the 「Auto」 checkbox): when on, the text colour switches to white or ink automatically based on background brightness (whichever contrasts better). On by default; picking a colour manually turns it off.
Align the selected text layer to the left. Position X is the left edge.
Centre-align the selected text layer. The default. Position X is the centre point.
Align the selected text layer to the right. Position X is the right edge.
Letter spacing slider with a number box, -2 to 20 px, default 2. Collapsed inside "Advanced: text details" in the text editor.
Line height slider with a number box, 1 to 2.5. Defaults: main title 1.25, subtitle 1.4. Collapsed inside "Advanced: text details".
Toggle: add a drop shadow to the selected text layer. On for the main title by default, off for the subtitle and badge. Collapsed inside "Advanced: text details".
Toggle: add an outline to the selected text layer. Off by default. White text gets an ink outline, ink text gets a white outline. Collapsed inside "Advanced: text details".
Position X number box for the selected text layer, 0 to 100 (percent of canvas width). It updates in sync when you drag the text on the canvas.
Position Y number box for the selected text layer, 0 to 100 (percent of canvas height). It updates in sync when you drag the text on the canvas.
Click the Logo upload area or drag an image in. PNG, JPG, and WebP are supported, with a single image no larger than 20MB. After upload the same area turns into a file card; click or drag a new file to replace it directly.
Logo layer width slider with a number box, 5 to 30 (percent of canvas width), default 12.
Position X number box for the Logo layer, 0 to 100 (percent of canvas width).
Position Y number box for the Logo layer, 0 to 100 (percent of canvas height), default 86.
Toggle under the canvas: draw a dashed 5% margin guide on the preview canvas. On by default. Xiaohongshu 3:4 also shows top and bottom margin warning bands. Guides are never drawn on export.
The 「⛶ Enlarge」 button under the canvas: opens a large preview where you can drag text or the logo to adjust its position. Close with Esc, the ✕ button, or by clicking the backdrop.
The input for AI title expansion, at the top of the left panel. Write just a title, or paste a title plus copy for the AI to use as reference, up to 500 characters. The box is a multi-line text area.
AI title style dropdown: tech geek / warm and healing / strict business / eye-catching viral. The chosen style sets the wording of the generated titles.
Generate a main title hook, a subtitle, and style tags from your topic and chosen style, and apply them to the canvas with one click. Requires login, 1 point per run. When logged out the button is greyed out; clicking it opens the login dialog first. The button shows the per-run cost; balance stays visible under the button.
Export format dropdown: PNG, JPG, or WebP (smaller file). JPG has no transparency; a hidden background exports as white.
Export scale dropdown: 1x or 2x supersampling, default 2x.
Export the cover in the current format and scale and download it. The file name looks like micacover_xiaohongshu_20261109-1730.png.
Copy the cover image to the clipboard (always PNG, the most clipboard-friendly format) so you can paste it straight into Xiaohongshu, WeChat, or another editor. If the browser does not support it, the status line suggests downloading instead.
The checkbox next to 「Save & share」, checked by default. When checked, saving also publishes the cover and creates a public link, and the icon share panel appears. When unchecked, the cover is saved privately only — no link is created and no share panel appears; you can publish it later from your account's Cloud Storage tab.
Export the current canvas in the chosen format and scale and save it to your cloud storage (source recorded as micacover, counts towards the 100MB quota). With 「Publish」 checked (the default), it is also published with a public link; unchecked, it is saved privately only. If you are not logged in, the login dialog opens first. If the storage is full, the status line offers two choices: “Clear all & retry” and “Cancel save”.
Only appears when the storage is full on save: keeps published files and clears the rest by default, uncheck to clear everything, then retries the save once; the result shows in the status line.
Only appears when the storage is full on save: dismisses the choice without clearing or saving.
The read-only link in the share panel after the cover is saved and published. Anyone opening it sees the cover image and title without logging in. Click the copy icon to copy it.
The X icon button in the share panel. It opens X's compose box in a new window with the text (title plus description) pre-filled; the cover shows up as a link. One click jumps there — nothing is posted automatically.
The Pinterest icon button in the share panel. It opens Pinterest's create-Pin page in a new window with the cover image and description pre-filled together. One click jumps there — nothing is posted automatically.
The Facebook icon button in the share panel. It opens Facebook's share dialog in a new window with only the public link — the title and image come from the link page. Facebook does not support pre-filled copy or direct image upload. One click jumps there — nothing is posted automatically.
The LinkedIn icon button in the share panel. It opens LinkedIn's share dialog in a new window with only the public link — the title and image come from the link page. Pre-filled copy is not supported. One click jumps there — nothing is posted automatically.
The 「Copy caption」 button in the share panel. It copies the current canvas's main title, subtitle, and tag (joined as a #tag) in one click, ready to paste into the compose box of Instagram, Xiaohongshu, WeChat, or any platform without a web sharing entry.
The 「Unpublish」 button in the share panel. The public link stops working immediately — opening it again shows a not-found page. The cover itself stays in your cloud storage and can be viewed or re-published from your account's Cloud Storage tab.
Drag text or the Logo directly on the canvas to move it. The X / Y number boxes update in sync, and you can also use them for precise alignment. The align buttons in the layer list align all layers with one click.
Click text or the Logo on the canvas to select that layer. The matching row in the list on the right highlights, and the layer controls appear under the canvas.
The 👁️ / 🚫 button at the start of each layer row. Click to hide or show the layer. Hidden layers are not drawn and not exported.
The ↑ button on each layer row. Moves the layer one step up (drawn later, on top). The background and gradient overlay layers are locked and cannot move.
The ↓ button on each layer row. Moves the layer one step down. The background and gradient overlay layers are locked and cannot move.
The red ✕ at the end of each layer row. Click once to delete the layer, no confirmation. The background and gradient overlay layers cannot be deleted.
A deleted layer can be added back: "+ Add main title / subtitle / badge / Logo" buttons appear under the layer list (only for the types currently missing). Click one to add it back with default styling at its original position and select it. The background and gradient overlay layers cannot be deleted, so they never need adding.
The 「Left align」 button in the layer list. It snaps the left edges of all movable layers (main title, subtitle, badge, Logo) to the 5% safe margin in one click, leaving their vertical positions untouched. The background and gradient overlay layers are locked and do not move.
The 「Centre horizontally」 button in the layer list. It centres every movable layer horizontally (each layer's centre on the canvas centre) in one click, leaving vertical positions untouched. The background and gradient overlay layers do not move.
The 「Right align」 button in the layer list. It snaps the right edges of all movable layers to the 95% line in one click, leaving their vertical positions untouched. The background and gradient overlay layers do not move.
The 「Centre vertically」 button in the layer list. It moves all movable layers together as one group, up or down, so the group's vertical centre lands on the canvas centre, keeping the spacing between layers unchanged. The background and gradient overlay layers do not move.
Click a row on the right to select that layer. The matching element on the canvas can then be dragged, and its controls appear under the canvas.
Export the current config as a .json file that can be imported later to restore it.
Pick a previously exported .json config file to restore. If the file is invalid or an unsupported version, the status line says so and the current canvas is left untouched.
How to Let AI Use It
Follow the steps to operate the browser page — no API key needed
There is no MCP tool for this one. Let an AI assistant follow these steps to operate the browser page (button IDs and labels match the workbench above):
- Open the tool page https://tools.waoww.com/tools/micacover/
- Pick a platform size from the size preset dropdown (#mc-preset) — 14 sizes, Xiaohongshu 3:4 is 1242×1660. Pick a gradient, a solid colour, or an image for the background on the left. For an image, upload a local file in the list upload row, or click 「Get a stock photo」 (#mc-stock-open) to pick one from the Unsplash / Pexels library (sign-in required; when it opens, the AI analyses the theme and suggests English search keywords — the top word fills the search box #mc-stock-q and searches once, the suggested keywords below it search again with one click; pick a photo and click 「Use this photo」 #mc-stock-confirm). The gradient overlay dropdown (#mc-overlay-gradient) turns the overlay off when you choose 「Do not enable」.
- Click the main title row in the layer list on the right, then type the title in the text box under the canvas (#mc-text). Font size (#mc-size-num) and position X / Y (#mc-x / #mc-y) all take typed numbers, or drag directly on the canvas. The align buttons in the layer list (#mc-layer-align-left / #mc-layer-align-center-h / #mc-layer-align-right / #mc-layer-align-center-v) align all movable layers with one click.
- To let AI write the title, type a topic in the theme box on the left and click 「AI title expansion」 (#mc-ai-btn). Requires login, 1 point per run; the result applies to the canvas automatically.
- Pick PNG / JPG / WebP in the export format dropdown (#mc-export-format) and 1x / 2x in the scale dropdown (#mc-export-scale), then click 「Download」 (#mc-download) to export. Check the status in #mc-status and the summary in #mc-summary.
- To share to social media, click 「Save & share」 (#mc-save-cloud). The cover is saved to your cloud storage; the 「Publish」 checkbox (#mc-publish) is checked by default, and when checked the cover is published too and the share panel shows a public link (#mc-share-url). Then click the X (#mc-share-x), Pinterest (#mc-share-pinterest), Facebook (#mc-share-facebook), or LinkedIn (#mc-share-linkedin) icon to jump to that platform. When unchecked, the cover is saved privately only and no share panel appears. 「Copy caption」 (#mc-copy-caption) copies the title, description, and tags, and 「Unpublish」 (#mc-unpublish) makes the public link stop working immediately.
Example prompt (send it to your AI assistant as-is)
「Open tools.waoww.com/tools/micacover/ and make a Xiaohongshu 3:4 cover: main title 「Get an export quote done in 3 days」, subtitle 「Template included」, ink gradient overlay, export JPG 2x」
How It Works & Highlights
How it works, and where it fits
This tool composes cover images locally on the Canvas: the background is a linear gradient generated live on the Canvas (with a deterministic noise texture), a solid colour, an uploaded image, or a stock photo picked from the Unsplash / Pexels library (searched through this site's proxy — API keys live only on the server; the library is for signed-in users only; when the picker opens, the AI first analyses the theme and suggests the best English search keywords — the top word fills the search box and searches once, the alternatives sit under the box and search again with one click, each analysis costs 1 point and is free with BYOK, and the same theme is only analysed once per page session; when the AI fails or the theme is empty, the picker falls back to a query built from the main title and main-title / subtitle suggestions; each source returns 10 photos at a time and 「All」 shows 20). Cover fills the canvas or contain shows the whole image, plus position, blur, and dim controls. On top sits a switchable gradient overlay that darkens the background to set off the text. The main title, subtitle, badge, and Logo are separate layers, drawn bottom to top in list order. Text switches between black and white automatically based on background brightness (you can override it by hand) and supports wrapping, letter spacing, and line height. Text and the Logo can be dragged directly on the canvas, and every position also has X / Y number boxes for precise adjustment. The align buttons in the layer list align all movable layers with one click (left / right snap to the 5% safe margin, centre horizontally, or centre the whole group vertically). Config changes autosave to this browser and are restored automatically when you reopen the page (a locally uploaded background image or Logo is not cached — the background falls back to a gradient on restore; a stock photo is saved by URL only and fetched again on restore). Export re-draws to an offscreen Canvas at the chosen scale (1x / 2x) without the safe zone guides, then encodes PNG, JPG, or WebP for download or clipboard copy. Clicking 「Save & share」 uploads the exported cover to your cloud storage (counts towards the 100MB quota) and, with the 「Publish」 checkbox checked (the default), publishes it, creating a public link page with the title, description, and cover image for X, Pinterest, Facebook, and LinkedIn to read when sharing; unpublishing makes the link stop working immediately. AI title expansion only sends your topic (title plus copy, up to 500 characters) to this site's AI gateway — the image itself is never uploaded.
Highlights
- ✓ Fourteen platform sizes in a dropdown: square, Xiaohongshu 3:4, WeChat header, plus Facebook / Instagram / X / LinkedIn / YouTube / Pinterest sizes. Switching sizes scales the fonts and keeps the positions
- ✓ Eight built-in gradients generated live, with a light noise texture. Solid colour or an uploaded background image also work, with fit, position, blur, and dim controls
- ✓ Layered layout: main title, subtitle, badge, and Logo are separate layers with visibility toggles, up/down ordering, and one-click delete. Drag on the canvas or use the X / Y number boxes
- ✓ Auto contrast: text switches between black and white based on background brightness, with a manual override. Full text controls for size, weight, letter spacing, line height, shadow, and stroke
- ✓ Safe zone guides: a dashed 5% margin line always on, plus top and bottom margin warning bands for Xiaohongshu 3:4. Guides never appear on export
- ✓ AI title expansion: four styles (tech geek / warm and healing / strict business / eye-catching viral). One run returns a main title hook, a subtitle, and style tags, for 1 point
Tips
- 💡 Keep the main title to 12 characters or fewer and the font size maxed out, so it reads at a glance in a phone feed
- 💡 Dark gradients go with white text and light gradients with ink text. Leave auto contrast on and it recalculates whenever the background changes
- 💡 On Xiaohongshu covers, keep key information out of the top and bottom warning bands — the feed crops and covers those areas
Notes
- ⚠️ A public link is a long random ID by default (not guessable), with no password and no expiry. 「Save & share」 publishes by default; uncheck 「Publish」 to save privately only, and publish later from your account's Cloud Storage tab. Unpublishing a published file makes its link stop working immediately.
- ⚠️ Clicking 「Save & share」 uploads the exported cover to your cloud storage and counts towards the 100MB quota. Making the cover itself stays fully local in the browser — nothing is uploaded unless you save.
- ⚠️ Facebook and LinkedIn do not support pre-filled copy: sharing only sends the public link, and the title and image come from the link page. Instagram, Xiaohongshu, and WeChat have no web sharing entry — download the cover and use 「Copy caption」 to paste the text into the app.
- ⚠️ The config autosaves to this browser (preset, text, colours, gradient, layout); the background image and Logo are not cached. Switching devices or clearing browser data loses the config, and on restore the background falls back to a gradient and images need re-uploading. Use 「Export config .json」 to carry a config to another device.
- ⚠️ The align buttons in the layer list act on all movable layers at once — main title, subtitle, badge, and Logo (the background and gradient overlay stay put). Left align snaps to the 5% safe margin, right align to the 95% line, horizontal centre puts each layer on the canvas centre, and vertical positions stay put; vertical centre moves the whole group of layers together, so the spacing between layers does not change.
- ⚠️ Stock photos come from Unsplash and Pexels and are free to use. When you publish a cover, please credit the photographer (every photo in the picker shows its photographer's name underneath). Search goes through this site's proxy; API keys live only on the server and never pass through the browser. The library is for signed-in users only, and so is its search API.
FAQ
Privacy, format limits, and usage tips
Related tutorials
Step-by-step guides for this tool on our blog