- COST
- USAGE
- BURDEN
- HUMAN
Short answer
How can I compare energy costs across different ZIP codes?
Your AI pulls billing metrics and usage data from PG&E datasets to map out cost differences. You'll see exactly where rates and consumption patterns diverge between specific areas. This breakdown shows how the data flows from the dataset to your analysis.
Analysis outcomes
What you'll discover
The data you pull results in these specific insights.
COST
Regional price maps
The AI aggregates billing data to show where rates are highest. You get a clear view of price disparities between ZIP codes.
USAGE
Consumption patterns
It compares how much electricity different areas use. This helps you see if high costs are driven by usage or base rates.
BURDEN
Affordability gaps
Your agent calculates energy spending relative to local demographics. You'll see which communities face the heaviest energy burdens.
HUMAN
Policy direction
The AI provides the raw comparisons, but you decide how to use these findings for community support or local policy.
The workflow
What your AI does when the data arrives.
The AI handles the heavy lifting of querying and comparing large datasets.
Fetch raw metrics
The agent pulls specific billing and usage records for the ZIP codes you specify.
get_billing_dataCompare regions
It runs a comparison between the datasets to find differences in cost and consumption.
compare_regionsIdentify trends
The AI looks for patterns in the billing data to highlight outliers or rising costs.
get_billing_dataSummarize findings
It organizes the comparison results into a clear summary for your review.
compare_regions
Try it
Copy these to start.
Use these prompts to begin your analysis.
Starting points
Swap the bracketed ZIP codes for the specific areas you want to study.
PG&E Public Datasets Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_jU4H9PpngMJSWTMWiUpgHMUI2qIOmttvQ6Nykwgk/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — PG&E Public Datasets capabilities are ready to use.
{
"mcpServers": {
"pge-public-datasets-mcp": {
"url": "https://edge.vinkius.com/vk_preview_jU4H9PpngMJSWTMWiUpgHMUI2qIOmttvQ6Nykwgk/mcp"
}
}
}Compare the average billing metrics for ZIP code 94102 and 94105.
Get the usage patterns for the 94103 area to see how they differ from surrounding regions.
Show me a comparison of energy affordability between these three ZIP codes: 94107, 94108, and 94109.
Identify which of these ZIP codes has the highest electricity cost per kilowatt hour based on the billing data.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect PG&E Public Datasets once, then ask.
One connection gives your AI access to the datasets. Your credentials stay encrypted and secure on our end.
Connect PG&E Public Datasets to your AIFAQ
How this task behaves
- 01
Can the AI change my actual PG&E billing account settings?
No. This MCP only reads public datasets for analysis; it cannot access or modify personal accounts.
- 02
Can I use this to predict next month's bill?
No. The tools are designed to compare and retrieve existing historical data, not to run predictive forecasting models.
- 03
Does the AI provide the raw CSV files?
The AI processes the data to answer your questions, but it does not provide direct file downloads.
- 04
Can I compare data from utility companies other than PG&E?
No. This specific Connector is limited to the PG&E Public Datasets provided in the manifest.
- 05
Will the AI automatically send reports to my email?
No. The AI provides the analysis within your chat interface. You must manually copy or export any findings you want to share.
More questions about PG&E Public Datasets? The Connector page answers them. See everything the PG&E Public Datasets Connector can do
Connect PG&E Public Datasets to Claude, Cursor, ChatGPT & more
