QuizSolver AI

• The cursor changes to a crosshair while selecting.
• The selection is automatically sent to the configured AI provider.
• The answer appears in a floating popup directly over the page.
• Option to directly download the capture as a PNG image.

🧠 MULTI-AI PROVIDER ───────────────────── Supports four artificial intelligence providers, each with selectable models:

┌─────────────────────────────────────────────────────────────────────────┐ │ 🔵 Google Gemini │ │ Models: gemini-2.5-flash · gemini-2.5-flash-lite · gemini-2.5-pro │ │ gemini-3-flash-preview · gemini-3.1-pro-preview │ │ gemini-3.1-flash-lite-preview │ │ │ │ 🟢 OpenAI ChatGPT │ │ Models: gpt-4.1-nano · gpt-4.1-mini · gpt-4.1 · gpt-4o-mini │ │ gpt-4o · gpt-5.4-nano · gpt-5.4-mini · gpt-5-mini · o4-mini│ │ │ │ 🟣 Venice AI │ │ Models: qwen3-vl-235b-a22b · mistral-small-3-2-24b-instruct │ │ google-gemma-3-27b-it · google.gemma-4-26b-a4b-it │ │ google.gemma-4-31b-it │ │ │ │ 🟠 Anthropic Claude │ │ Models: claude-sonnet-4-6 · claude-opus-4-6 · claude-haiku-4-5 │ └─────────────────────────────────────────────────────────────────────────┘

Each provider has its own API Key field and selectable model. Direct links to each provider's console to obtain the key and add credits.

🎯 CUSTOM PROMPTS ──────────────────

• Create your own prompts with a title and content for quick reuse.
• Full CRUD system: create, edit, duplicate and delete prompts.
• Select any saved prompt to send it along with the capture.
• Default prompts per provider: configure a different prompt for

Gemini, ChatGPT, Venice and Claude independently.

🕵️ BYPASS SYSTEM (Anti-Detection) ─────────────────────────────────── Scripts that neutralize the most common detection and restriction mechanisms on online exam platforms:

🔓 Anti-Focus (Canvas LMS): Neutralizes tab-change or focus-loss detection.

• Blocks focus, blur and visibilitychange events.
• Spoofs document.visibilityState = 'visible' and document.hidden = false.
• Ideal for Canvas LMS that pauses the exam when you leave the tab.

🖱️ Anti-Context Menu (Right Click): Restores the right-click context menu on sites that block it.

• Intercepts events in the capturing phase.
• Neutralizes preventDefault() for contextmenu, copy, cut, paste, dragstart.
• Allows text selection and copying on sites that disable it.

🔧 Anti-DevTools (Developer Console): Neutralizes all DevTools detection methods.

• Function.prototype.toString spoofing to hide proxies.
• Fakes outerWidth/outerHeight dimensions to conceal the panel.
• Smooths timing of Date.now() and performance.now() during debugger pauses.
• Removes debugger statements in eval(), Function(), setTimeout, setInterval.
• Neutralizes external detection libraries (devtools-detect, Firebug).

🖱️ Anti-MouseLeave (Cursor Always Present): Makes the site believe the cursor never leaves the window.

• Blocks mouseleave, mouseout, pointerleave, pointerout events only

when the cursor leaves the viewport (does not break menus or tooltips).

• document.hasFocus() always returns true.
• Compatible with MutationObserver for dynamically added elements.

📋 PDF GENERATOR (Canvas LMS) ────────────────────────────── On Canvas LMS quiz pages (*/courses/*/quizzes/*) a script is automatically activated that generates a complete PDF of the exam with all questions and options for offline review.

📁 CAPTURE HISTORY ───────────────────

• Stores all captures with date and time.
• Thumbnails of each capture with preview.
• Button to view the full capture in a new tab (dedicated viewer).
• Download any individual capture as PNG.
• Delete captures individually or clear the entire history.
• Unlimited storage using chrome.storage with IndexedDB.

💾 EXPORT / IMPORT CONFIGURATION ──────────────────────────────────

• Export all your settings (API keys, prompts, preferences) to a

downloadable JSON file as a backup.

• Import settings from a JSON file to restore or share configuration

between devices.

🌐 BROWSER CONTEXT MENU ──────────────────────── Options accessible by right-clicking on any page to activate extension functions without opening the popup.

🌍 MULTI-LANGUAGE SUPPORT ────────────────────────── The extension is available in 10 languages: Spanish 🇪🇸 · English 🇬🇧 · Portuguese (Brazil) 🇧🇷 · Portuguese (Portugal) 🇵🇹 German 🇩🇪 · French 🇫🇷 · Italian 🇮🇹 · Japanese 🇯🇵 · Norwegian 🇳🇴 · Swedish 🇸🇪

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 🗂️ POPUP SECTION DESCRIPTION ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🏠 Main Panel (main) ───────────────────── The popup home screen. From here you can:

• Enable/disable each of the 4 bypasses with switches.
• Select the active AI provider (Gemini / ChatGPT / Venice / Claude).
• Expand each provider panel to enter the API Key,

select the desired model and write the session prompt.

• Click "Capture Question" to start area selection.
• Access all other panels from the header buttons.

✏️ Prompts Panel (prompts) ─────────────────────────── Full custom prompts manager:

• Creation form with title field and content textarea.
• Table with all saved prompts.
• Edit ✏️, duplicate 📋 and delete 🗑️ buttons per row.
• Prompts are saved in chrome.storage.local persistently.

⭐ Default Prompts Panel (defaultPrompts) ────────────────────────────────────────── Allows assigning a different base prompt for each AI provider. The default prompt is sent automatically when no session prompt is selected.

🕐 History Panel (historial) ───────────────────────────── Chronological list of all captures with:

• Thumbnail of the captured image.
• Exact date and time of capture.
• View 👁️, download ⬇️ and delete 🗑️ buttons per capture.
• Button to clear the entire history at once.
• Access to the full Viewer that opens captures in a new tab.

📤 Export/Import Panel (export) ────────────────────────────────

• "Download" button → generates and downloads the config.json file.
• "Import" button → opens the file picker to load a JSON file.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 📖 MINI TUTORIAL — HOW TO USE QUIZSOLVER AI ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

════════════════════════════════════ STEP 1 — 🔧 Installation ════════════════════════════════════

1. Download or build the extension (npm run build). 2. Open Chrome and go to: chrome://extensions/ 3. Enable "Developer mode" (top right corner). 4. Click "Load unpacked" and select the /dist folder. 5. The QuizSolver AI icon will appear in the toolbar. 📌 Recommended: pin it by clicking the pin 📌 in the extensions manager.

════════════════════════════════════ STEP 2 — 🔑 Configure your API Key ════════════════════════════════════

1. Click the extension icon in the Chrome toolbar. ✅ The Main Panel popup opens. 2. Choose your preferred AI provider using the radio buttons: ○ Gemini ○ ChatGPT ○ Venice ○ Claude 3. Click on the provider name to expand its panel. 4. Paste your API Key in the corresponding field. 💡 Use the "Get API KEY" link to go directly to the provider console if you don't have one yet. 5. Select the model you want to use from the dropdown. 6. (Optional) Write a session prompt or select a saved one. 7. Configuration is saved automatically as you type.

════════════════════════════════════ STEP 3 — 📸 Capture a Question ════════════════════════════════════

1. Navigate to the exam or quiz on any webpage. 2. Open the extension popup. 3. Click the 📷 "Capture Question" button. → The popup closes and the cursor changes to a crosshair ✛ 4. Click and hold on the top-left corner of the question, drag to include all options and release. → A selection rectangle is drawn as you drag. 5. On release, the capture is processed and sent to the AI provider. 6. The answer appears in a floating popup over the page! 🎉 → You can also view the capture in history to review it.

════════════════════════════════════ STEP 4 — 🕵️ Activate Bypasses ════════════════════════════════════

From the Main Panel, activate switches as needed:

┌──────────────────────────────────────────────────────────────────────┐ │ 🔓 Anti Focus (Canvas LMS) │ │ Enable if the exam is on Canvas LMS and pauses when switching │ │ tabs. Prevents the site from detecting you left the page. │ │ │ │ 🖱️ Unlock context menu │ │ Enable if the site blocked right-click or text selection. │ │ You can copy, paste and use the menu normally. │ │ │ │ 🔧 Unlock developer console │ │ Enable if the site detects and blocks the browser DevTools. │ │ Neutralizes all known detection methods. │ │ │ │ 👻 Simulate cursor always on site │ │ Enable if the exam pauses or detects when you move the cursor │ │ outside the browser window. │ └──────────────────────────────────────────────────────────────────────┘

⚠️ Bypasses activate immediately on the current page. Some may require reloading the exam tab (F5).

════════════════════════════════════ STEP 5 — ✏️ Create Custom Prompts ════════════════════════════════════

1. In the popup, click the ✏️ "Manage Prompts" icon. 2. Write a short title (e.g. "Concise multiple choice"). 3. Write the full prompt in the textarea. E.g.: "You are an expert. Analyze the image and respond ONLY with the letter of the correct option. No explanations." 4. Click "Save". 5. The prompt will appear in the table and be available for selection in the Main Panel when making a capture. 6. You can duplicate 📋 existing prompts to modify them easily.

════════════════════════════════════ STEP 6 — 📁 Review History ════════════════════════════════════

1. Open the popup and click the 🕐 "Capture History" icon. 2. You'll see a list of all captures with thumbnails and dates. 3. Click 👁️ "View" to open the capture in full screen in a new tab. 4. Click ⬇️ to download the capture as a PNG file. 5. Click 🗑️ to delete an individual capture. 6. Use "Clear all" to wipe the entire history.

════════════════════════════════════ STEP 7 — 💾 Export your Configuration ════════════════════════════════════

1. Open the popup and click ⬇️ "Export / Import". 2. Click "Download" to save all your configuration to a JSON file (API keys, prompts, selected model, etc.). 3. To restore on another device: click "Import" and select the previously downloaded JSON file.

🔒 Keep your backup file in a safe place as it contains your API Keys.

════════════════════════════════════ STEP 8 — 📄 Generate PDF (Canvas LMS) ════════════════════════════════════

1. Navigate to a Canvas LMS quiz: (URL like: https://university.instructure.com/courses/.../quizzes/...) 2. The extension automatically detects the URL and activates the generator. 3. A "Generate PDF" button will appear on the quiz page. 4. Click it to download a PDF with all exam questions and options for offline review.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ⚙️ REQUIRED PERMISSIONS ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

• activeTab → Capture screenshot of the active tab.
• tabs → Create new tabs (capture viewer).
• storage → Save settings, prompts and history.
• unlimitedStorage → Store capture history without limits.
• scripting → Inject bypass scripts into pages.
• contextMenus → Add options to the browser context menu.
• downloads → Download PNG captures and generated PDFs.
• host_permissions → Work on any URL for captures and bypasses.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 📝 FINAL NOTES ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

• The extension requires a valid API Key with credits from the chosen provider

to process captures.

• Captures are processed directly from the browser to the provider's API.

They are not sent to any external server.

• The capture history is saved locally in your browser.
• Version: 1.1.2 | Chrome name: QuizSolver AI

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━