Find answers or browse by topic

Select one or more frames you want feedback on and click the eye button (UX Feedback) in the sidebar. Leave the prompt empty for a general review, or add a specific ask (e.g. 'focus on hierarchy and spacing'). In AI Settings → Vision, choose what to analyze. Run the analysis and iterate from the results panel.

Visual Design covers layout, spacing, typography, and color usage. UX & Psychology reviews clarity, affordances, cognitive load, and user flow. Brand Communication evaluates alignment with tone and visual identity. Accessibility highlights contrast, sizing, and inclusive design hints. Personas & Needs sketches who the design is for and what they need, from your Document Context or the design itself. Creative Ideas explores alternative directions and quick experiments.

Yes. Click the + button next to the text box and choose Screenshot selection, Upload image, or Upload file. Images can be PNG, JPEG, WebP or any other image type your browser reads (up to 4 per message; Contently shrinks them to 2048 px). Text files can be TXT, MD, CSV, JSON, XML, HTML, CSS, JS or TS. You can also paste images directly into the text box.

Smart Intent is the default mode in the Brainstorm panel. As you type, the AI reads your message and automatically picks the most suitable skill for it. You can click the active skill chip anytime to browse skills manually or return to Smart Intent.

Right-click a group header and choose ✨ Layer rename. For image groups choose 'Match Image' to rename layers from what the image depicts, 'Description' for descriptive names, or 'Tag Style' for short Title Case tags (up to 8, e.g. 'Mountain Sunset Hiking'). Other groups offer layout naming styles instead: Design System, Slash Paths, BEM, or Code Names.

Leave it on 'Auto (best balance)' for a model that balances quality and cost. To pin one, pick a specific model from the model list in AI Settings → API Key.

AI providers retire older models regularly. Contently checks against your provider once a day and refreshes the list, falling back to the closest equivalent model. Click 'Check for new models' in AI Settings → API Key → Advanced options to refresh immediately.

Add a short sentence in the component description field explaining its purpose. Name components clearly (e.g., Button, Checkbox, Card). Expose real component properties (text, boolean, variant) rather than deeply nested unlinked text. Prefix scratch components with '_' (or '.') to exclude them.

Add Document Context (AI Settings → Context) when the AI needs stable background information such as target audience, copy tone of voice, product principles, or technical constraints. It is saved in your Figma file, per document (up to 8,000 characters), and applied across requests.

No. AI feedback provides suggestions and highlights potential issues. It does not replace verifying contrast ratios against WCAG requirements or testing with real assistive technology.

Select text layers in your file that represent realistic content. Click the + Template button and choose 'Create Text Template'. Name it and save: the plugin stores all selected text values as template items. Tip: use the same name as your Figma layers to enable Smart Template auto-matching.

Image Templates are curated sets of images. Create one from selected image layers or the Images panel. Select target layers and click Apply. Use 'Apply to → Smart' to auto-match by layer name, or 'Direct selection' for manual control.

Icon Templates let you browse and apply icon sets to selected layers, including remote libraries like Material Symbols by Google. Search by name, narrow with the variant chips, toggle Filled, and click an icon to replace the selection, or use Apply All for Smart Template auto-replacement.

Material Symbols is a built-in remote library: no external Figma file needed. Open the icon templates, choose Material Symbols by Google, search by name (3 or more characters), and toggle Filled. Icons are applied as Figma components.

Controls how template items are assigned: Random (shuffled each time, ideal for mockups), Original (preserves the exact list order), or Ascending/Descending (alphabetical or numerical sorting).

Matching layer names first tries Smart Template name-matching before applying to other selections. Entire selection applies directly across all selected layers. Replace text options include Whole text, Only numbers (e.g. Order #100 -> Order #873), or Only text (keeps numbers intact).

Smart matches by layer name first, then falls back to image fills. Matching smart templates only replaces layers with matching names. Direct selection only applies to whatever is selected. Image fills only replaces fills in shapes and frames, ignoring layer names.

Modifiers let you control template application right from the layer name. Syntax: template-name:modifier1:modifier2. Modes include :numbers (replaces only numbers), :text (replaces only words), :upper/:lower/:title (case formatting), :1/:2 (fixed item index), and conditionals like :if-numbers, :if-text, :if-empty.

Inline placeholders mix dynamic template values with static copy inside a single text layer using the {TemplateName} syntax (e.g., 'Order #{OrderIDs} for {CustomerNames}' or 'Hello {Names:upper}!'). To insert one, open a text template's detail view and click the 'Insert inline placeholder' button: it adds {Name} to the selected text layer.

Click the Menu button in the sidebar and open Compress Images. Set a target maximum file size (default 1.0 MB, up to 2 MB), choose whether to compress the source images or individual images ('Compress Source Image'), and run. Contently shrinks image file weight while preserving visual sharpness on canvas.

Yes. A public Figma Community file with sample frames, text, and images lets you practice Smart Templates, modifiers, AI feedback, and image compression safely.

Open AI Settings → API Key tab. Choose a provider (OpenAI, Groq, Gemini, or Local) and enter your credentials. For Local (LM Studio / Ollama), no API key is needed: expand Advanced options, enter the Base URL (e.g. http://localhost:1234/v1) and the Model, and click 'Save API Settings'.

For most designer workstations: Qwen2.5-VL-7B-Instruct. It provides sharp layout and color perception, handles design screenshots reliably, and runs smoothly on consumer hardware with 4-bit or 8-bit quantization.

1. Install LM Studio from lmstudio.ai. 2. Search and load a model (e.g. Qwen2.5-VL). 3. Switch to the Developer tab and start the server. 4. Enable 'Enable CORS' in Developer Settings. 5. In Contently, point Base URL to http://localhost:1234/v1.

1. Install from ollama.com. 2. Set environment variable OLLAMA_ORIGINS=* to permit Figma iframe connections. 3. Run 'ollama serve'. 4. Pull a model (e.g. 'ollama pull llama3.2-vision'). 5. In Contently, set Base URL to http://localhost:11434/v1.

Figma plugins run in sandboxed iframes. If your local server rejects cross-origin requests, a CORS error appears. In LM Studio, toggle 'Enable CORS' in Developer Settings. In Ollama, set OLLAMA_ORIGINS=* and restart.

Verify the server process is running and that no firewall is blocking ports 1234 or 11434. Figma only allows localhost, so use http://localhost:… and not 127.0.0.1 or another machine's address. In AI Settings → API Key → Advanced options, click 'Auto-detect' then 'Test' to check connectivity.

In LM Studio, ensure a model is actively loaded in memory before starting the server. In Ollama, verify that you have pulled at least one model ('ollama list'). You can also type the model identifier manually into the Model field in Contently.

No. With a local provider Contently sends AI requests only to your local server on localhost, so prompts, copy and screenshots are not sent to a cloud AI provider. The plugin's own Local card says the same: your prompts, content and API keys never leave your device. Contently still contacts raiken.io for license checks and, if you allow it, usage tracking (Menu → System Settings → Privacy Settings).

Contently is a Figma plugin by David Zerbe (RAIKEN) for editing text and image content, reusing templates, and working with repeated layouts. AI assistance is optional.

Yes. Start with the free content editing and template tools. The free plan includes 12 AI actions per day, shared by text, vision and image; paid upgrades unlock advanced tools and unlimited local operations. Third-party AI providers may charge separately for their API usage.

No. You can edit and fill content, reuse templates, organize layers, and work with Smart Overflow completely offline without configuring an AI provider.

No. The plans are one-time plugin purchases for lifetime access with no recurring subscription fees.

Contently supports OpenAI, Google Gemini, Groq, and any OpenAI-compatible local endpoint on localhost (LM Studio, Ollama). Model availability depends on your configured credentials.

Depending on the action, only the prompt, selected text, document context, or screenshots you explicitly include are sent to your chosen provider. Review your provider's terms before sending sensitive designs.

Click the Menu button in the sidebar, choose Premium Status & Activation, paste the license code from your purchase confirmation and click Add.

Three categories, listed under Menu → System Settings → Privacy Settings. Essential (always on) records plugin start and stop and critical errors. Anonymous Analytics counts which features are used. Performance reports slow or failed operations. The last two are optional and switched on by default; Reject Optional turns them off. A license check sends your license key, email and Figma user id to raiken.io so the license can be verified.

Only inside component instances. 'Include hidden layers' makes Contently look inside hidden layers of component instances. Layers that are hidden themselves are never matched in Find Similar or Manipulate, and the toggle has nothing to do with off-canvas layers.

In the Refine tab, type a word and press Enter or comma to create a search chip. Click the operator badge between chips to combine them with OR (matches either), AND (matches both), or NOT (matches the first but excludes the second). Chips are evaluated strictly left to right, and a match is a case-insensitive part of a layer name or its text.

The Manipulate tab lets you reshape an active selection without starting over. Invert selects the siblings of your current selection instead. Select Every Nth keeps only every Nth sibling (e.g. every 2nd card for alternating patterns); it is on by default, so switch it off to use Invert.

Controls how the content list organizes layers: By Layer Name (strict name-based grouping), By Parent Name (clusters by parent frame or component), By Element Type (groups all text, shapes, etc.; the default), or By Element Name. It is in the content toolbar.

Live Select keeps the plugin's content list continuously in sync with your canvas selection. It is off by default. On very large documents with thousands of layers, leave it off and click 'Scan Selection' to update the list in batches without repeated canvas rescans.

Smart Templates automatically match template data to the right layers based on layer names, so you don't have to select each layer manually. When you click Apply Matching Templates, the plugin scans your selection and fills each layer named after a template: text, image, and icon templates all work this way.

1. Create a template (text, image or icon) and give it a clear name (e.g., 'Product Names'). 2. In Figma, name the target layer(s) after it: 'Product Names'. 3. Select the parent frame or any layers in your design. 4. Click 'Apply Matching Templates' to fill every matching layer across nested components, or open one template and click 'Apply All' to fill only the layers named after it.

Yes. Select your text layers and apply a template. The plugin batches all updates, loads fonts automatically, and preserves layer structure. For the fastest workflow, use Smart Templates so the plugin auto-matches by layer name.

Smart Overflow belongs to Copy-Paste to List. When your text has more lines than the layout has elements, Contently creates new elements for the extra lines and tells you ('N lines couldn't fit. Creating new elements...'). Start with a short list and review text wrapping, layout and component structure before applying it to a larger design.