Oculo
@xidik12
About Oculo
AI-powered native browser with built-in MCP server. 7 tools for browser automation — navigate, click, fill forms, extract data, run pipelines. ~30 tokens per page description vs 25,000 for DOM snapshots. macOS Apple Silicon. MIT licensed.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"oculo": {
"command": "node",
"args": [
"~/Desktop/oculo/bin/oculo-mcp.mjs"
]
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Oculo?
Oculo is an AI-powered native browser built with Electron that integrates with MCP clients to let AI assistants see and interact with web pages. It works with Claude Code, Cursor, Windsurf, Zed, or any MCP-compatible client.
How to use Oculo?
Connect Oculo to any stdio MCP client by pointing it to node /path/to/mcp/oculo-mcp.mjs. For Claude Code, run claude mcp add oculo -- node /path/to/mcp/oculo-mcp.mjs. For Cursor, add the command and args to .cursor/mcp.json.
Key features of Oculo
- AI Chat Panel with multi-provider support
- 7 MCP tools (~30 tokens per page description)
- Media generation (images and videos)
- Smart Auth: credential vault, auto-login, TOTP 2FA
- Privacy-first: 4-tier permission gate, PII redaction
- Voice control via on-device Whisper STT
- Sandboxed browser-side JavaScript execution
Use cases of Oculo
- Automate web tasks through AI-driven clicking and typing
- Extract structured data (lists, tables, cards) from pages
- Fill forms intelligently using label matching
- Generate images/videos from text prompts
- Run multi-step browser pipelines in a single call
FAQ from Oculo
What MCP tools does Oculo provide?
It provides seven tools: page (describe page, ~30 tokens), act (navigate/click/type, ~10 tokens), fill (fill form fields, ~15 tokens), read (extract data, ~100 tokens), run (multi-step pipeline, ~200 tokens), media (generate images/videos, varies), and shell (execute JS, ~50 tokens).
How do I connect Oculo to Claude Code?
Run claude mcp add oculo -- node /path/to/mcp/oculo-mcp.mjs in your terminal.
What platforms are supported?
Only macOS Apple Silicon is currently supported. Windows and Linux versions are coming soon.
What privacy features does Oculo have?
Oculo includes a 4-tier permission gate, PII redaction, anti-injection measures, and an audit log.
What is the token cost for describing a page?
The page tool uses approximately 30 tokens to describe a page’s headings, forms, buttons, and links.
Frequently asked questions
What MCP tools does Oculo provide?
It provides seven tools: `page` (describe page, ~30 tokens), `act` (navigate/click/type, ~10 tokens), `fill` (fill form fields, ~15 tokens), `read` (extract data, ~100 tokens), `run` (multi-step pipeline, ~200 tokens), `media` (generate images/videos, varies), and `shell` (execute JS, ~50 tokens).
How do I connect Oculo to Claude Code?
Run `claude mcp add oculo -- node /path/to/mcp/oculo-mcp.mjs` in your terminal.
What platforms are supported?
Only macOS Apple Silicon is currently supported. Windows and Linux versions are coming soon.
What privacy features does Oculo have?
Oculo includes a 4-tier permission gate, PII redaction, anti-injection measures, and an audit log.
What is the token cost for describing a page?
The `page` tool uses approximately 30 tokens to describe a page’s headings, forms, buttons, and links.
Basic information
More Other MCP servers
Maestro
mobile-dev-incPainless E2E Automation for Mobile and Web
Production-ready MCP integrations for AI applications
Klavis-AIKlavis AI: MCP integration platforms that let AI agents use tools reliably at any scale
🚀 Model Context Protocol (MCP) Curriculum for Beginners
microsoftThis open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,
Website
FunnyWolfAdversary simulation and Red teaming platform with AI
FastMCP v2 🚀
jlowin🚀 The fast, Pythonic way to build MCP servers and clients.
Comments