ocular MCP Server

by xyun1996 Your server? Claim it

local only

ocular runs on your own machine — the client starts it, so there is no endpoint to ping. 20 installs a week from npm. Last commit 15 Oct 2026.

Vision tools for coding agents: screenshots, OCR, UI diffs, errors, tables, and charts.

Installs per day peak 11 · avg 4 · -33% w/w

a month agotoday

Installs / week

npm · ocular-mcp

Stars

2 open issues

15 Oct 2026

Last commit

1 release in 90 days

MIT

License

TypeScript

Nothing serious here today

Today is the operative word: we check ocular every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.

Connect this server

This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.

Claude Code Claude Desktop Codex Cursor VS Code

run in your terminal

claude mcp add ocular -- npx -y ocular-mcp

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "ocular": {
      "args": [
        "-y",
        "ocular-mcp"
      ],
      "command": "npx"
    }
  }
}

~/.codex/config.toml

[mcp_servers.ocular]
command = "npx"
args = ["-y", "ocular-mcp"]

.cursor/mcp.json

{
  "mcpServers": {
    "ocular": {
      "args": [
        "-y",
        "ocular-mcp"
      ],
      "command": "npx"
    }
  }
}

.vscode/mcp.json

{
  "mcpServers": {
    "ocular": {
      "args": [
        "-y",
        "ocular-mcp"
      ],
      "command": "npx"
    }
  }
}

This one needs environment variables set before it will start: OCULAR_BASE_URL (OpenAI-compatible API base URL), OCULAR_API_KEY (API key for the configured vision model provider), OCULAR_MODEL (Vision-capable model identifier). The author declared them in the registry entry; get the values from the project itself.

Alternatives to ocular

same job, measured the same way

One Shot UI

by tn0123

Deterministic UI extraction and screenshot diffing for AI coding agents.

67 installs/wk local only

Visily UI Design

by visily-app

UI design from prompts, screenshots, and URLs for AI coding agents and theme tokens.

answering

Figma Context MCP Air

by compassalessandrolorenz

Figma MCP server for AI agents. 9 tools: design context, variables, screenshots, and FigJam.

24 installs/wk local only

Ramus Android

by ramus

Disposable cloud Android emulators for coding agents: run an APK or PR build, tap, type, screenshot.

answering

Agent Tools UK

by lexyholdcraft-source

Pay-per-use agent tools: screenshots, web to Markdown, PDF tables, OCR, UK company checks

10 tools answering

Mobile Device MCP

by saranshbamania

AI control of Android/iOS devices. Screenshots, UI tree, AI vision, Flutter, video. 49 tools.

96 installs/wk local only

Hwatu

by hongnoul

Verification browser for coding agents: headless WebKit, DOM eval, screenshots, pixel-diff.

local only

Anylens

by slowper

Turn videos, PDFs, screenshots, and articles into anchored, cached understanding for agents.

60 installs/wk local only

ocular — questions

Answers built from our own checks of this server.

Why is there no uptime for ocular?

ocular runs on your own machine over stdio — there is no network address to reach, so uptime cannot be measured for it by anyone. What can be measured is adoption: the npm package ocular-mcp was installed 20 times last week.

How do I connect ocular?

Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It runs locally, so the command pulls ocular-mcp straight from npm; nothing to host, nothing to sign up for.

How many people use ocular?

The npm package ocular-mcp was installed 20 times in the last week. Week over week that is -33%. We show installs rather than GitHub stars on purpose: a star is a bookmark, an install is someone actually running it.

Is ocular open source?

Yes — it is published under the MIT licence, written in TypeScript, 0 stars on GitHub and 2 open issues. The source link is on this page, so you can read exactly what it does with your data before you connect it.