Tekuila

Provides access to Tekuila restaurant menus with AI-powered healthy meal recommendations via MCP clients.
  • python

0

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

You can access Tekuila restaurant menus and receive AI-powered healthy meal recommendations through this MCP server. It supports daily and weekly menus, AI analysis, and date-aware planning, making it easy to integrate with your MCP clients for fast, health-focused meal planning.

How to use

Connect with your MCP client to fetch Tekuila menus and get intelligent health recommendations. You can query today’s menu, request this week’s planning, or ask for AI-assisted analyses of menu options. The server offers two transports: a lightweight local stdio transport for development and a cloud-ready HTTP transport for hosting. Use the stdio transport for local development and MCP client integration, or run the HTTP transport to expose the service over HTTP.

How to install

Prerequisites: Python 3.12 or higher and the uv runtime you use to run MCP servers.

Step 1: Clone the project and set up the local directory.

git clone <your-repo>
cd tekuila
uv sync

Step 2: Run the server in your preferred transport.

# Cloud hosting (HTTP transport)
uv run python main.py

# Local development (stdio transport)
uv run python tekuila.py

Server connections and transport options

The server supports two transport modes. Use the HTTP transport for cloud hosting and the stdio transport for local development. The HTTP transport exposes an HTTP server on port 8000, while the stdio transport communicates via standard input/output with MCP clients.

HTTP transport (remote server) configuration example (URL is provided when running locally):

{
  "type": "http",
  "name": "tekuila_http",
  "url": "http://127.0.0.1:8000",
  "args": []
}

Stdio transport (local development) configuration example (runs via uv with the tekuila.py script):

{
  "type": "stdio",
  "name": "tekuila_stdio",
  "command": "uv",
  "args": ["run", "python", "tekuila.py"]
}

Connect to MCP clients

You can wire the Tekuila MCP server into clients like Claude Desktop, Cursor, and LM Studio using the stdio transport. The example below shows configuring Claude Desktop to launch the stdio-based Tekuila server from a local directory.

{
  "mcpServers": {
    "tekuila": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/tekuila",
        "run",
        "tekuila.py"
      ]
    }
  }
}

Available tools and prompts

The server exposes functions to retrieve menus and plan meals, plus AI analysis capabilities to help you choose healthier options.

# Example function names exposed by the server
get_current_day_menu()
get_current_week_menu()
get_current_date()
analyze_daily_menu()
plan_weekly_menu()

Menu structure and AI analysis

Menu categories are presented in Finnish with health-focused prioritization. Primary options include Vegaaninen kasvislounas (vegan lunch), Kasvislounas (vegetarian lunch), Lounas (regular lunch with meat), POP UP Bistro (special/expensive options), and Jälkiruoka (desserts). The AI analysis assesses nutritional quality, processing level, taste balance, and overall value, while flagging overly processed items and heavy sauces.

Usage examples

Daily Menu Analysis: Ask for today’s menu and AI guidance on the healthiest choices.

Weekly Planning: Request a weekly meal plan with healthy recommendations.

Troubleshooting and notes

If the server does not start, verify you’re using Python 3.12+ and that dependencies are installed via uv sync. Confirm the RSS feed URLs are accessible if you rely on external data feeds.

Development and project layout

The project includes entries for stdio and HTTP transports, along with a Python-based MCP server implementation and an HTTP entry point for cloud hosting.

Available tools

get_current_day_menu

Fetch today's menu with current date context.

get_current_week_menu

Fetch this week's menu with date context.

get_current_date

Retrieve the current date and time context.

analyze_daily_menu

Analyze today's menu with AI analysis instructions.

plan_weekly_menu

Generate a weekly planning guide with healthy recommendations.

analyze_menu_selection

AI-powered analysis of a specific menu selection.

weekly_menu_planning

AI-powered weekly meal planning tool.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational