uthere
Camera-based presence detection MCP server for Claude Code. Uses your webcam to check if you're sitting at your computer, and can automatically keep Claude working when you step away.
README
u there?
Camera-based presence detection for Claude Code. Uses your webcam to check if you're sitting at your computer.
Two components:
- MCP server exposes an
is_user_presenttool that Claude can call anytime to check if you're there - Stop hook automatically keeps Claude working when you step away, so it doesn't stop and wait for input from an empty chair
Uses MediaPipe BlazeFace for fast, lightweight face detection. Works well with glasses, varied lighting, and different angles. The model (~200KB) is downloaded automatically on first run.
Install
Requires Python 3.11+ and uv.
git clone https://github.com/odfalik/uthere.git
cd uthere
uv sync
Camera permissions
On macOS, your terminal app needs camera access. Go to System Settings > Privacy & Security > Camera and enable it for your terminal (Terminal, iTerm2, etc).
Setup
Add either or both of the following to your Claude Code settings file (~/.claude/settings.json).
MCP server
Gives Claude a tool it can call to check if you're at your desk.
{
"mcpServers": {
"uthere": {
"command": "uv",
"args": ["--directory", "/path/to/uthere", "run", "python", "-m", "uthere.server"]
}
}
}
Stop hook
Prevents Claude from stopping when you're away. If Claude finishes a response and no face is detected, it gets a message telling it to keep going. A 60-second cooldown prevents infinite loops.
{
"hooks": {
"Stop": [
{
"hooks": [
{
"type": "command",
"command": "uv --directory /path/to/uthere run python -m uthere.hook",
"timeout": 15
}
]
}
]
}
}
Configuration
All configuration is through environment variables. Set them in your shell profile or in Claude Code's env settings.
| Variable | Default | Description |
|---|---|---|
UTHERE_AWAY_MESSAGE |
"No face detected. User appears to be away..." |
MCP tool message when no face is detected |
UTHERE_PRESENT_MESSAGE |
"User is at their computer." |
MCP tool message when face is detected |
UTHERE_HOOK_MESSAGE |
"The user is not at their computer right now..." |
Message sent to the agent when the stop hook blocks |
Example using Claude Code's env settings:
{
"env": {
"UTHERE_AWAY_MESSAGE": "User is away. Send them a message on Discord.",
"UTHERE_HOOK_MESSAGE": "User is AFK. Keep working, ping them on Discord if stuck."
}
}
How it works
- Opens the webcam via OpenCV
- Takes a few warmup frames (lets auto-exposure settle)
- Checks 3 consecutive frames with BlazeFace face detection
- Returns
trueif a face is found in any frame
The whole check takes about 1-2 seconds. The camera is released immediately after each check.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.