> **New to Claude?** Start with [INSTANT_START_WITH_CLAUDE.md](./INSTANT_START_WITH_CLAUDE.md) for a guided tutorial on using Claude for Geotab development. This page is a quick reference for specific prompts.
Sign in to like and favorite skills
New to Claude? Start with INSTANT_START_WITH_CLAUDE.md for a guided tutorial on using Claude for Geotab development. This page is a quick reference for specific prompts.
This is your control center. Copy these prompts into Claude, ChatGPT, or your AI assistant to build Geotab apps instantly.
[!CAUTION] NEVER paste real production credentials into an AI chat. If using a real fleet database, ask the AI to write code that loads credentials from a
file. Only paste passwords if you are using a disposable demo database..env
Goal: Connect to your fleet and see what's possible.
I want to explore the Geotab API and build something cool. Database: [Insert Database Name] Username: [Insert Email] # SECURITY NOTE: For production, assume credentials are in a .env file. # Only if this is a DEMO account: Password: [Insert Password] Server: my.geotab.com 1. Connect to my fleet. 2. Show me a summary of my vehicles and recent trips. 3. Suggest 3 simple project ideas I could build right now based on this data.
Goal: Understand your specific data patterns.
Look at the actual data from my fleet (trips, exceptions, engine status) for the last 7 days. 1. Identify the top 3 most common events or behaviors. 2. Find one outlier or anomaly (e.g., a very long trip, high idling). 3. Create a simple ASCII chart or table summarizing fleet activity by day.
Goal: Generate a working script for a specific idea.
Let's build the [PROJECT IDEA NAME] we discussed. Requirements: 1. Write a complete, single-file Python script. 2. Use the `mygeotab` library. 3. Include error handling for authentication and API limits. 4. Add clear comments explaining how to run it. 5. Save the output to a CSV file named 'fleet_report.csv'. Do not ask clarification questions yet, just give me the best v1 implementation.
Goal: Turn your script into a visual web app.
Take the logic we just wrote and convert it into a simple web dashboard. Stack: - Python (Flask or Streamlit) - Chart.js for visualization Features: 1. A map showing vehicle locations (if lat/lon available). 2. A table of the data. 3. Auto-refresh every 60 seconds. Give me the complete code in one file (if Streamlit) or the necessary files.
Goal: Debug errors when they happen.
I am getting this error: [PASTE ERROR MESSAGE HERE] 1. Explain why this is happening. 2. Fix the code to handle this case gracefully. 3. Explain how to prevent it in the future.
Create a Python script to fetch all devices (vehicles) from my account and display their Name, SerialNumber, and ID in a table.
Fetch 'StatusData' for fuel levels for the last 24 hours. Calculate the total fuel consumed per vehicle and Identify the least efficient vehicle.
Fetch 'ExceptionEvent' data for speeding and harsh braking from the last week. Rank drivers by detailed safety score (100 = perfect).
Fetch all Trip data for the last 30 days and export it to a clean CSV file with columns: Vehicle, StartTime, EndTime, Distance, MaxSpeed.