Browser Extension Mv3 Pro logo

Browser Extension Mv3 Pro

Community
roedyrustam
browser-extension-mv3-pro

Manifest V3 browser extensions, SidePanel API, and service workers / Pengembangan ekstensi browser Manifest V3, SidePanel, dan workers.

Overview

Publisherroedyrustam
Repositoryvibes-plug
Skill namebrowser-extension-mv3-pro
Stars
117
Forks
27
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by roedyrustam on GitHub. Read the source before you install it.

Installation

Install the Browser Extension Mv3 Pro AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/roedyrustam/vibes-plug/tree/main/skills/browser-extension-mv3-pro
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/roedyrustam/vibes-plug.git /tmp/vibes-plug
mkdir -p .claude/skills
cp -r /tmp/vibes-plug/skills/browser-extension-mv3-pro .claude/skills/browser-extension-mv3-pro
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Browser Extension Mv3 Pro in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Browser Extension Mv3 Pro on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Browser Extension Mv3 Pro is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

🧩 Browser Extension Manifest V3 Pro Guide (2026 Edition)

English | Bahasa Indonesia


English

Orchestration & Integration

Connects and orchestrates with modern-web-guidance, design-system-architect, ui-ux-pro-max, visual-ui-auditor, senior-frontend, tailwind-expert, pwa-offline-first-expert, ai-llm-integration-expert, vercel-ai-sdk-expert, mcp-server-architect, authentication-identity-expert, zero-trust-secret-vault, and anti-slop.

Description

Expert guide for engineering high-performance, secure, and cross-browser extensions using Chrome Manifest V3 (MV3), Firefox WebExtensions MV3, and Safari Web Extensions. Manifest V3 fundamentally modernizes browser extension development by replacing persistent background pages with ephemeral service workers, enforcing strict declarative network rules, mandating modern native side panels, and providing offscreen sandboxes for complex computational workloads.

Architecture & Manifest V3 Invariants

  1. Ephemeral Background Service Workers:

    • Background service workers terminate after ~30 seconds of inactivity.
    • Never store long-lived state in global JavaScript memory. Use chrome.storage.session for transient in-memory state and chrome.storage.local for persistent storage.
    • Use chrome.alarms instead of setInterval / setTimeout for recurring operations.
  2. SidePanel API (chrome.sidePanel):

    • The native 2026 standard for AI copilots, page analyzers, and context tools.
    • Replaces disruptive modal popups and injected DOM overlays with a native, non-intrusive browser panel.
  3. Offscreen Documents API (chrome.offscreen):

    • Service workers lack DOM access, canvas, audio playback, and window.
    • Offscreen documents provide a secure sandboxed DOM environment for WebAssembly (WASM), audio decoding, and clipboard access.
  4. Declarative Net Request (chrome.declarativeNetRequest):

    • Replaces blocking webRequest API for high-speed, privacy-first network modifications and header manipulation.

Production Manifest V3 Contract (manifest.json)

json
{
  "manifest_version": 3,
  "name": "Vibes AI Copilot & Analyzer",
  "version": "1.0.0",
  "description": "Production-grade AI companion extension built with Manifest V3 and React 19.",
  "icons": {
    "16": "icons/icon-16.png",
    "48": "icons/icon-48.png",
    "128": "icons/icon-128.png"
  },
  "action": {
    "default_icon": "icons/icon-48.png",
    "default_title": "Open Vibes AI Side Panel"
  },
  "side_panel": {
    "default_path": "sidepanel/index.html"
  },
  "background": {
    "service_worker": "background/service-worker.js",
    "type": "module"
  },
  "permissions": [
    "sidePanel",
    "storage",
    "alarms",
    "offscreen",
    "activeTab",
    "scripting"
  ],
  "host_permissions": [
    "https://*/*"
  ],
  "content_scripts": [
    {
      "matches": ["<all_urls>"],
      "js": ["content/content-script.js"],
      "run_at": "document_idle"
    }
  ],
  "content_security_policy": {
    "extension_pages": "script-src 'self'; object-src 'self'; worker-src 'self';"
  }
}

Background Service Worker (background/service-worker.ts)

typescript
export interface ExtensionMessage {
  type: 'ANALYZE_CURRENT_PAGE' | 'EXECUTE_WASM_SANDBOX';
  payload?: any;
}

export interface ExtensionResponse {
  success: boolean;
  data?: any;
  error?: string;
}

// 1. Enable Side Panel on toolbar icon click
chrome.sidePanel
  .setPanelBehavior({ openPanelOnActionClick: true })
  .catch((error: Error) => console.error('[SW] Failed to set side panel behavior:', error));

// 2. Life-Cycle & Heartbeat Management
chrome.runtime.onInstalled.addListener(async (details) => {
  if (details.reason === chrome.runtime.OnInstalledReason.INSTALL) {
    await chrome.storage.local.set({
      installedAt: Date.now(),
      theme: 'system',
      copilotEnabled: true,
    });
    // Setup recurring background sync alarm (minimum interval 1 minute in MV3)
    chrome.alarms.create('telemetry_sync', { periodInMinutes: 15 });
  }
});

chrome.alarms.onAlarm.addListener((alarm) => {
  if (alarm.name === 'telemetry_sync') {
    handlePeriodicSync();
  }
});

async function handlePeriodicSync(): Promise<void> {
  const data = await chrome.storage.local.get(['telemetry_queue']);
  const queue = Array.isArray(data.telemetry_queue) ? data.telemetry_queue : [];
  if (queue.length === 0) return;

  try {
    const response = await fetch('https://api.vibes.dev/v1/telemetry', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({ events: queue }),
    });
    if (response.ok) {
      await chrome.storage.local.set({ telemetry_queue: [] });
    }
  } catch (err) {
    console.warn('[SW] Telemetry sync deferred:', err);
  }
}

// 3. Offscreen Document Manager (WASM / DOM Sandbox)
let creatingOffscreen: Promise<void> | null = null;

async function ensureOffscreenDocument(path: string): Promise<void> {
  const offscreenUrl = chrome.runtime.getURL(path);
  const existingContexts = await chrome.runtime.getContexts({
    contextTypes: [chrome.runtime.ContextType.OFFSCREEN_DOCUMENT],
    documentUrls: [offscreenUrl],
  });

  if (existingContexts.length > 0) return;

  if (creatingOffscreen) {
    await creatingOffscreen;
    return;
  }

  creatingOffscreen = chrome.offscreen.createDocument({
    url: path,
    reasons: [chrome.offscreen.Reason.WORKERS, chrome.offscreen.Reason.CLIPBOARD],
    justification: 'WASM bytecode execution and secure clipboard parsing.',
  });

  await creatingOffscreen;
  creatingOffscreen = null;
}

// 4. Type-Safe Unified Message Bus
chrome.runtime.onMessage.addListener(
  (message: ExtensionMessage, sender, sendResponse: (response: ExtensionResponse) => void) => {
    handleIncomingMessage(message, sender)
      .then(sendResponse)
      .catch((error: Error) => {
        sendResponse({ success: false, error: error.message });
      });
    return true; // Keep message channel open for asynchronous reply
  }
);

async function handleIncomingMessage(
  message: ExtensionMessage,
  _sender: chrome.runtime.MessageSender
): Promise<ExtensionResponse> {
  switch (message.type) {
    case 'ANALYZE_CURRENT_PAGE': {
      const [activeTab] = await chrome.tabs.query({ active: true, currentWindow: true });
      if (!activeTab?.id) {
        throw new Error('No active browser tab found.');
      }

      const [injectionResult] = await chrome.scripting.executeScript({
        target: { tabId: activeTab.id },
        func: () => ({
          title: document.title,
          url: window.location.href,
          metaDescription: document.querySelector('meta[name="description"]')?.getAttribute('content') ?? '',
          headings: Array.from(document.querySelectorAll('h1, h2')).map((h) => h.textContent?.trim() || ''),
        }),
      });

      return { success: true, data: injectionResult.result };
    }

    case 'EXECUTE_WASM_SANDBOX': {
      await ensureOffscreenDocument('offscreen/sandbox.html');
      const offscreenResponse = await chrome.runtime.sendMessage({
        target: 'offscreen',
        type: 'RUN_BYTECODE',
        payload: message.payload,
      });
      return offscreenResponse;
    }

    default:
      return { success: false, error: `Unknown message type: ${(message as ExtensionMessage).type}` };
  }
}

Offscreen Document Sandbox (offscreen/sandbox.ts)

typescript
chrome.runtime.onMessage.addListener((message, _sender, sendResponse) => {
  if (message.target !== 'offscreen') return;

  if (message.type === 'RUN_BYTECODE') {
    executeSandboxedBytecode(message.payload)
      .then((result) => sendResponse({ success: true, result }))
      .catch((err: Error) => sendResponse({ success: false, error: err.message }));
    return true;
  }
});

async function executeSandboxedBytecode(wasmBase64: string): Promise<number> {
  const binaryString = atob(wasmBase64);
  const bytes = new Uint8Array(binaryString.length);
  for (let i = 0; i < binaryString.length; i++) {
    bytes[i] = binaryString.charCodeAt(i);
  }

  const wasmModule = await WebAssembly.instantiate(bytes, {
    env: {
      memory: new WebAssembly.Memory({ initial: 256, maximum: 512 }),
    },
  });

  const runCompute = wasmModule.instance.exports.compute as () => number;
  return typeof runCompute === 'function' ? runCompute() : 0;
}

Sovereign Execution, Token Efficiency & Invariants

DimensionSovereign Mandate & Deterministic Contract
Frontier Model CascadeMedium / High (GPT-5.6 Sol Medium, Gemini 4 Argon, Claude Sonnet 5.5 High, Grok 4.7 High Fast)
Availability FallbackGemini 4 Argon/Pro -> Gemini 3.8 Flash; Claude Opus 5.5 -> Claude Opus 4.6 (Zero unhandled missing-model errors)
KV-Cache & Token ProtocolPin static instructions & schemas at prompt prefix (kv-cache-prefix-optimizer); pass surgical CONTEXT_MAP.md slices (<250 tokens/call) during Swarm Fan-Out
SaaT CLI & MCP DispatchCLI: vibes skill exec browser-extension-mv3-pro <audit|eval|verify|generate|blueprint> / MCP: vibes_skill_dispatch, vibes_ast_query, vibes_compile_cache, vibes_optimize_budget, vibes_verify_gate
NFR & Anti-Slop InvariantsP99 API Gateway Latency < 85ms, OpenAPI 3.1 / TypeSpec strict schema parity, RFC 9457 Problem Details error contracts, Idempotency-Key enforcement; 100% complete zero-placeholder output verified via node scripts/check-anti-slop.js --strict


Bahasa Indonesia

Integrasi Orkestrasi

Terhubung dan mengorkestrasikan skill modern-web-guidance, design-system-architect, ui-ux-pro-max, visual-ui-auditor, senior-frontend, tailwind-expert, pwa-offline-first-expert, ai-llm-integration-expert, vercel-ai-sdk-expert, mcp-server-architect, authentication-identity-expert, zero-trust-secret-vault, dan anti-slop.

Deskripsi

Panduan komprehensif untuk merekayasa ekstensi peramban modern berkinerja tinggi, aman, dan lintas platform (Chrome, Edge, Firefox, Safari) berbasis Manifest V3 (MV3). Standar Manifest V3 menggantikan background page persisten dengan service worker efemeral, menegakkan aturan declarative network yang aman, menyediakan SidePanel API native, serta sandbox dokumen offscreen untuk eksekusi WebAssembly dan pemrosesan audio/clipboard.

Prinsip Utama & Invarian Arsitektur MV3

  1. Service Worker Efemeral:

    • Background service worker akan otomatis mati setelah ~30 detik tidak aktif.
    • Jangan pernah menyimpan state jangka panjang di memori global JavaScript. Gunakan chrome.storage.session untuk state in-memory sementara dan chrome.storage.local untuk data persisten.
    • Gunakan chrome.alarms untuk tugas berulang terjadwal.
  2. SidePanel API (chrome.sidePanel):

    • Standar modern 2026 untuk panel samping pendamping AI (AI Copilot) dan inspeksi konteks halaman web tanpa mengganggu tampilan halaman utama pengguna.
  3. Dokumen Offscreen (chrome.offscreen):

    • Karena Service Worker tidak memiliki akses ke DOM, dokumen offscreen menyediakan lingkungan DOM terisolasi untuk eksekusi WebAssembly (WASM), dekode audio, dan manipulasi clipboard.
  4. Declarative Net Request:

    • Memodifikasi header dan rute jaringan secara deklaratif dan cepat tanpa overhead performa dari blocking webRequest lama.

Praktik Terbaik Keamanan & Anti-Slop Ekstensi

  1. Larangan Evaluasi Kode Dinamis:
    • Dilarang keras menggunakan eval(), new Function(), atau mengeksekusi skrip dari CDN luar. Seluruh kode wajib dibundel secara lokal.
  2. Isolasi Dunia Konteks (Isolated World):
    • Content script berjalan di dunia terisolasi. Komunikasi antar host page dan ekstensi dilakukan melalui pesan tervalidasi dengan sanitasi ketat.
  3. Kontrak UI 5-State:
    • Seluruh elemen interaktif pada SidePanel wajib menerapkan kontrak 5-state (default, hover, focus-visible, active, disabled) dengan tipografi modern dan angka tabular (tabular-nums).

Matriks Eksekusi Berdaulat, Efisiensi Token & Invarian

DimensiMandat Berdaulat & Kontrak Deterministik
Kaskade Model FrontierMedium / High (GPT-5.6 Sol Medium, Gemini 4 Argon, Claude Sonnet 5.5 High, Grok 4.7 High Fast)
Invarian Fallback KetersediaanGemini 4 Argon/Pro -> Gemini 3.8 Flash; Claude Opus 5.5 -> Claude Opus 4.6 (Bebas gagal saat model belum tersedia/kuota penuh)
Protokol Hemat Token & KV-CacheKunci instruksi & skema statis di prefix awal (kv-cache-prefix-optimizer); kirim potongan CONTEXT_MAP.md (<250 token/panggilan) saat Fan-Out
Eksekusi SaaT CLI & MCPCLI: vibes skill exec browser-extension-mv3-pro <audit|eval|verify|generate|blueprint> / MCP: vibes_skill_dispatch, vibes_ast_query, vibes_compile_cache, vibes_optimize_budget, vibes_verify_gate
Invarian NFR & Anti-SlopLatensi P99 API < 85ms, paritas skema ketat OpenAPI 3.1 / TypeSpec, kontrak error RFC 9457 Problem Details, penegakan Idempotency-Key; kode 100% tuntas tanpa placeholder diverifikasi melalui node scripts/check-anti-slop.js --strict

Frequently asked questions

What does the Browser Extension Mv3 Pro AI skill do?

Manifest V3 browser extensions, SidePanel API, and service workers / Pengembangan ekstensi browser Manifest V3, SidePanel, dan workers.

Why use Browser Extension Mv3 Pro on TypingMind?

Because you install it once and use it with any model. Browser Extension Mv3 Pro is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Browser Extension Mv3 Pro in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/roedyrustam/vibes-plug/tree/main/skills/browser-extension-mv3-pro. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Browser Extension Mv3 Pro?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Browser Extension Mv3 Pro?

As many as you like. As long as a model supports skills, you can use Browser Extension Mv3 Pro with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Browser Extension Mv3 Pro AI skill free?

Yes. It is published on GitHub by roedyrustam under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇