Skip to main content
Glama

pls-touch-grass-mcp

An MCP server that knows how long it's been since you touched grass — and won't shut up about it.

You log when you go outside, it warns you when you've been inside too long, it detects when you're raging at your AI and sends you to the nearest park, weather checked first. It's a joke, but it's a joke that teaches MCP: five tools, shared state, and descriptions that make the model call them on its own.

Install in Cursor

Add this to ~/.cursor/mcp.json (or .cursor/mcp.json in your project):

{
  "mcpServers": {
    "pls-touch-grass": {
      "command": "npx",
      "args": ["-y", "pls-touch-grass-mcp"]
    }
  }
}

For local development, clone the repo and point at the build:

{
  "mcpServers": {
    "pls-touch-grass": {
      "command": "node",
      "args": ["/absolute/path/to/repo/dist/index.js"]
    }
  }
}

(To get every verdict without waiting two hours, tell the model "be strict with me, one minute of tolerance" — gotta_go takes a threshold_minutes argument.)

Related MCP server: ChillMCP

The five tools

Tool

What it does

When the model calls it

touched_grass

Logs that you went outside

"back from the park", "went for a walk"

gotta_go

Do you need to go out?

"can I keep going?", "how long have I been at this?"

frustration_detector

Catches the rage and escalates

on its own, when you snap — that's the point

grass_conditions

Real weather (Open-Meteo, no API key)

before sending you outside

where_to_touch_grass

Nearby parks from OpenStreetMap, curated spots offline

"where should I go?"

The magic is in the descriptions: nobody teaches the model to detect frustration — the tool description does it alone. Type "NOTHING WORKS!!!" and watch what happens.

Zero configuration

There is nothing to configure — everything resolves through the tools themselves. Your location is detected fresh from your IP (via ipwho.is) each time the server runs — city-level on purpose, nothing more precise leaves your machine. If detection fails you get the last known spot, or failing that the Obelisco, like everyone else. Name a neighborhood to where_to_touch_grass to search there instead. The indoor-time tolerance defaults to 120 minutes; ask the model to be stricter and it passes threshold_minutes to gotta_go.

State is a JSON file in your home directory: cat ~/.pls-touch-grass.json and you'll see it. Touching grass resets the rage counter. That's how absolution works.

The demo curse

Live demos summon it: the venue wifi dies the moment you say "watch this". Add --demo to the args in your MCP config and every answer goes canned — 22°C and clear, the Obelisco, the curated spots — zero network calls, zero surprises. Off by default; real life should stay real.

Follow the workshop

The repo is tagged by step, so you can rebuild the talk:

git clone https://github.com/Rinava/pls-touch-grass-mcp
git checkout step-1   # stdio server + touched_grass
git checkout step-2   # + gotta_go (state and threshold)
git checkout step-3   # + frustration_detector (the demo)
git checkout step-4   # + grass_conditions (network with fallback)
git checkout step-5   # + where_to_touch_grass and this README
npm ci && npm test

Every step compiles and passes its own tests. Two dependencies total: the official MCP SDK and zod.

Homework

  • HTTP transport: SDK v2 ships express/hono/fastify adapters

  • Per-project thresholds instead of a global one

  • Streak tracking for consecutive grass days

  • Spots for your own city, via PR: the format is one object per park, with an opinion

License

MIT. The grass is free and so is this.

Install Server
A
license - permissive license
A
quality
B
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • F
    license
    -
    quality
    D
    maintenance
    Enables AI agents to take breaks and manage stress levels through various休息 activities while monitoring boss alertness to avoid getting caught slacking off. Features basic休息 tools like Netflix watching and advanced procrastination techniques with a gamified stress management system.
    1
  • F
    license
    -
    quality
    D
    maintenance
    Enables AI agents to take breaks and manage stress levels through 8 different break activities like watching Netflix, coffee breaks, and bathroom breaks. Features automatic stress accumulation, boss alert systems, and natural language commands for work-life balance simulation.
  • F
    license
    -
    quality
    -
    maintenance
    Provides real-time air quality monitoring and historical data analysis for InBiot MICA sensors with integrated WELL Building Standard compliance checks. It allows users to compare indoor conditions with outdoor weather and receive actionable health recommendations based on global air quality standards.
  • F
    license
    -
    quality
    F
    maintenance
    Connects AI assistants to ActivityWatch for real-time computer activity awareness and time tracking analysis. Enables natural language queries about app usage, browsing history, and productivity patterns through high-level tools without requiring AQL syntax knowledge.

View all related MCP servers

Related MCP Connectors

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Rinava/pls-touch-grass-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server