How can we help?
Start with the guides below. Most questions are answered here.
Install & update
Step-by-step ZIP installation, system requirements, and how updates work.
Accounts & sign in
Create an account, log in, and understand what happens when you sign in on the site or in the extension.
Plans & limits
What Lite, Pro and Ultra include - images, chars, keys, library and video access.
Everything the extension does
What each feature does, how to use it, and which plan unlocks it - the exact rules the extension enforces.
Image prompt extractionLite · Pro · Ultra
What it does: Hover any image on any page and extract the exact prompt that produced it - subject, light, lens, style and technical detail.
How to use: Hover the image, click Extract Prompt, wait for analysis, copy the result.
Input → Output: Image → structured prompt. Modes: Standard (Lite+), Deep (Pro+), Ultra (Ultra).
Limits: 1 image per request on Lite · 1,000 / 2,000 / 20,000 characters per prompt by plan.
Prompt editor & chatAll plans
What it does: Refine an extracted prompt in a chat - change text, change style, change design direction, then reuse it in any AI model.
How to use: Open the editor next to an extraction and type your instructions in plain language.
Input → Output: Extracted prompt + instruction → revised prompt.
Limits: Prompt length by plan (1,000 / 2,000 / 20,000 chars).
Deep analysis (Pro)Pro · Ultra
What it does: Reads the design system - layout, typography, hierarchy, composition - and restructures the prompt for professional redesign rather than simple copying.
How to use: Select the Deep mode before extracting.
Input → Output: Design → visual-system analysis → restructured prompt.
Plans: Pro and Ultra only.
Ultra pipelineUltra
What it does: The Visual DNA engine. The image is analyzed through 14 structured sections - canvas forensics, object inventory, spatial geometry, subject/pose, face, hair, clothing and materials, camera and lighting reconstruction, color DNA with approximate hex palette, typography with occlusion, graphic elements, layer stack and relationship map, style and hierarchy - then compiled into a six-part document (reconstruction summary, Visual DNA blueprint, spatial layout map, layer/occlusion map, master reconstruction prompt, generator-specific versions for GPT Image / Midjourney / FLUX / Ideogram) and verified by a reconstruction gap audit.
How to use: Select the Ultra mode before extracting.
Input → Output: Image → 14 Visual DNA sections → six-part document (master prompt up to 20,000 chars + generator versions).
Plans: Ultra only.
Prompt libraryAll plans
What it does: Save prompts locally and reuse them instantly.
How to use: Click save on any extracted or edited prompt; browse saved items in the popup.
Limits: 3 saved items on Lite · 50 on Pro · 500 on Ultra.
History & saved chatsAll plans
What it does: Every extraction and chat stays in recent activity - local and private.
How to use: Open the Recent activity panel in the popup.
Plans: All plans; nothing is sent to the server.
Format converterPro · Ultra
What it does: Converts prompts between output formats in one click.
How to use: Open the converter and pick the target format for the current prompt.
Plans: Pro and Ultra.
Multiple API accounts1 / 3 / 10 keys
What it does: Add multiple OpenRouter / provider keys with per-account model selection and automatic failover.
How to use: Settings → Add account, pick models, save, then Test API.
Limits: 1 key on Lite · 3 on Pro · 10 on Ultra.
Video reverse engineeringMAX plan
What it does: Frame-based, segment-batched video analysis that turns a video into a production prompt with motion, scene and style detail - no download needed.
How to use: On any page (YouTube, Instagram, anywhere), hover the video and choose Extract Prompt. Pick a model (Spark Lite / Spark Pro), an aspect ratio (16:9 / 9:16 / 1:1) and a frame preset (1K / 2K / 4K), then run.
Access: MAX plan (and the 3-day free trial).
Updates & version historyAll plans
What it does: Shows the version changelog and checks the latest release. This channel updates by downloading the new release - the popup flags it when one is available.
How to use: Open the Updates button in the popup; if a new version exists, download it from the site and reload the extension.
Common questions
Organized by what you're trying to do.