Czech Transport Assistant by CzechFabric | Český dopravní asistent od CzechFabric
RemoteAbout Czech Transport Assistant by CzechFabric | Český dopravní asistent od CzechFabric
The official Python SDK for CzechFabric, which provides access to Czech public-service capabilities including transport and geolocation. It supports trip planning and real-time stop information, with caching, retries, and structured errors.
Key features
- 01Plan public transport trips
- 02Retrieve real-time departures
- 03Geocode and reverse-geocode locations
- 04Find nearby stops and stop details
- 05List server tools and prompt summaries
- 06Built-in caching, retries, and errors
Use cases
- 01Plan a public transport journey
- 02Check departures from a stop or coordinates
- 03Find nearby stops and inspect zones
- 04Review available tools for an LLM assistant
MCP support
The SDK connects to the CzechFabric MCP server and lets clients list its tools and generate prompt summaries. Requests require an API key.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.czechfabric.czclaude mcp add --transport http czechfabric_sdk https://mcp.czechfabric.cz{"mcpServers":{"czechfabric_sdk":{"type":"http","url":"https://mcp.czechfabric.cz"}}}Open in Cursor{"servers":{"czechfabric_sdk":{"type":"http","url":"https://mcp.czechfabric.cz"}}}Open in VS Code{"mcpServers":{"czechfabric_sdk":{"type":"http","url":"https://mcp.czechfabric.cz"}}}{"mcpServers":{"czechfabric_sdk":{"serverUrl":"https://mcp.czechfabric.cz"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://mcp.czechfabric.czAuth requiredTools
13| Parameter | Type | Required | Description |
|---|---|---|---|
| format | string | No | Response format: 'tone' (token-efficient TONE format, default), 'json' (standard JSON), or 'both' (includes both formats) |
| location | string | Yes | Location name, address, or landmark in Prague (e.g., 'Karlovo náměstí', 'Wenceslas Square', 'Anděl') |
| radius_meters | any | No | Search radius in meters around the location (default: 2000m = 2km) |
Details
From the project README
🇨🇿 CzechFabric SDK
The official Python SDK for accessing the Czech Intelligence Fabric (MCP) - a unified API for Czech public services including transport, geolocation, and contextual AI tooling.
🚀 Features
- ✅ Plan public transport trips using OpenTripPlanner
- 🚏 Real-time departures from stop names or coordinates
- 🗺️ Geocode / Reverse-geocode locations
- 📍 Find nearby stops
- 📦 Fetch stop metadata and zone info
- 🧠 List available tools and generate LLM-friendly summaries
- 🔁 Built-in retry, caching, and error handling
📦 Installation
pip install czechfabric-sdk
⚡ Quick Start
from czechfabric_sdk.client import CzechFabricClient
client = CzechFabricClient(
api_key="your-api-key",
base_url="https://mcp.czechfabric.cz"
)
# Example: Plan a trip
trip = await client.plan_trip("Florenc", "Karlovo náměstí", departure_time="in 15 minutes")
print(trip)
📚 Usage Examples
Explore detailed examples in examples.md:
- 🚉 Plan a trip
- 📆 Get departures
- 🧭 Geocode / Reverse geocode
- 📍 Nearby stops
- 🧾 Stop metadata
- 📋 List stops
- 🛠️ List available tools
- 🤖 Generate prompt summary
🧰 Tool Management
You can dynamically list tools exposed by the MCP server:
tools = await client.get_tool_names()
prompt = await client.get_tool_prompt_summary()
print(prompt)
This is helpful for LLM-based assistants or debugging available capabilities.
🔐 Authentication
All requests require an API key:
client = CzechFabricClient(api_key="your-api-key", base_url="https://mcp.czechfabric.cz")
You can generate your key from your CzechFabric user dashboard.
⚠️ Error Handling
The SDK provides structured exceptions:
NetworkError– server unreachable or timed outInvalidAPIKeyError– auth failureRateLimitExceededError– too many requestsToolExecutionError– tool execution failed
🧪 Running Tests
pip install -r requirements-dev.txt
pytest tests/
📎 License
MIT © 2025 Aliyu Abdulbasit Ayinde
🌐 Project Links
- 🧠 MCP Server: https://mcp.czechfabric.cz
- 💬 Feedback & Issues: GitHub Issues
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.