- LOAD PROFILES
- DEMAND SPAN
- ANOMALIES
- HUMAN REVIEW
Short answer
How can I analyze smart meter usage patterns with my AI?
Your AI uses UtilityAPI to fetch specific meter IDs and pull high-resolution interval data. It then processes these timestamps to build load profiles and spot demand spikes. You get a clear view of energy consumption behavior.
Analysis outcomes
What your agent produces
The AI turns raw utility data into actionable energy insights.
LOAD PROFILES
Detailed usage curves
The AI aggregates interval data into mathematical profiles. You see exactly when energy demand peaks and troughs occur.
DEMAND SPAN
Identify peak loads
Your agent flags specific time windows where consumption hits maximum levels. This helps you pinpoint heavy usage periods.
ANOMALIES
Spot unusual behavior
The AI compares current intervals against historical patterns. It alerts you to unexpected surges or drops in energy use.
HUMAN REVIEW
Validate findings
The AI presents the processed data and summaries. You decide which patterns require physical inspection or manual intervention.
The workflow
What your AI does when the data arrives.
The AI moves from discovery to deep data retrieval automatically.
Locate meters
The agent identifies all available meters associated with the account to establish the scope of the analysis.
list_metersFetch intervals
Once a meter is selected, the AI pulls the specific time-stamped consumption readings needed for the profile.
get_intervalsExtract data
The agent pulls the full set of meter information to ensure all necessary attributes are available for the model.
get_meter_dataPattern recognition
The AI analyzes the retrieved time-series data to find trends, peaks, and irregularities in the load.
get_intervals
Try it
Copy these to start.
Use these prompts to kick off your analysis.
Starting points
Replace the bracketed text with your specific meter IDs or account details.
UtilityAPI Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_UQcFN9jSbPr5jD0q3QcZWex8kd0ypw9wyAD09b4d/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 — UtilityAPI capabilities are ready to use.
{
"mcpServers": {
"utilityapi-mcp": {
"url": "https://edge.vinkius.com/vk_preview_UQcFN9jSbPr5jD0q3QcZWex8kd0ypw9wyAD09b4d/mcp"
}
}
}List all meters available in my UtilityAPI account so I can choose one to analyze.
Get the interval data for meter [ID] for the last 30 days and summarize the peak usage times.
Compare the load profiles for meter [ID] and meter [ID] to find differences in demand patterns.
Analyze the interval data from meter [ID] to identify any unusual energy spikes during nighttime hours.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect UtilityAPI once, then ask.
One click connects your preferred AI client to the UtilityAPI MCP. Your credentials stay encrypted and secure, and you can immediately start asking questions about your meter data.
Connect UtilityAPI to your AIFAQ
How this task behaves
- 01
Can my AI change my utility account settings?
No. The tools only allow the AI to read meter lists and retrieve usage data.
- 02
Does the AI store my utility login credentials?
No. Vinkius manages the connection securely; your AI client only interacts with the tools provided by the MCP.
- 03
Can the AI bill my utility provider directly?
No. There are no tools available to perform transactions or modify billing information.
- 04
How granular is the interval data?
The AI can retrieve data based on the specific intervals provided by the utility through the get_intervals tool.
- 05
Can the AI delete my meter history?
No. The AI has read-only access to the data provided by UtilityAPI.
More questions about UtilityAPI? The Connector page answers them. See everything the UtilityAPI Connector can do
Connect UtilityAPI to Claude, Cursor, ChatGPT & more
