Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for sustainability analysts

Site Energy Use Anomaly Agent

Find and close energy waste at sites within weeks, not at year end

MonitoringAnalysisScience and research
See how it worksGet access
Site Energy Use Anomaly Agent: what goes in, what the agent does and what you get

What it does

Energy efficiency work needs someone to notice when a building starts using more power than it should. Each week this agent pulls interval meter data or monthly bills for each site and weather degree days. It adjusts for weather and opening hours and compares use with each site's own baseline. When a site runs above the threshold for two periods, it classifies the pattern: night-time load, weekend use or a step change. It drafts a question for the site manager, which the analyst approves before sending. It logs the response and checks the next week's data to see whether use returned to baseline. Open issues are escalated after the set number of weeks. Savings are estimated only from measured data. The analyst approves any message to site leadership and any savings claim. Edge case: new equipment that raises use on purpose becomes a new baseline, not an anomaly.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
ApprovedYes, continueNo 1 STARTS WHEN Weekly data arrives 2 USES A TOOL Pull meter data and weather degree days 3 DOES Normalize and compare with each site's baseline 4 DOES Classify anomalies by pattern 5 YOU APPROVE Analyst approves questions to site managers 6 USES A TOOL Send question and log the response 7 CHECKS THE RESULT Did use return to baseline the following week? If not: keep the issue open, escalate after the setnumber of weeks. Back to step 3. 8 RESULT Weekly anomaly log with resolved and open issues
Read the steps as a list
  1. Weekly data arrives
  2. Pull meter data and weather degree days
  3. Normalize and compare with each site's baseline
  4. Classify anomalies by pattern
  5. Analyst approves questions to site managersThe agent waits here for your OK.
  6. Send question and log the response
  7. Did use return to baseline the following week?If not: keep the issue open, escalate after the set number of weeks. Back to step 3.
  8. Weekly anomaly log with resolved and open issues

How it decides

Use is normalized for weather and hours. A site above the threshold for two periods is flagged with the likely pattern behind it.

  • Flag after two periods above 10% of baseline
  • Night load above 40% of daytime load is a likely controls issue
  • Planned changes reset the baseline after analyst approval

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Anomaly threshold (default 10%)
  • Weeks before escalation
  • Sites in scope
  • Report day and format

What keeps you in control

It always asks you first

  • Messages to site managers
  • Resetting a baseline
  • Reporting savings

Hard limits

  • Never change building controls directly
  • Savings are reported only from measured data

It stops when

  • Done: all anomalies resolved or explained
  • Stop: meter data missing for a site for two weeks, ask facilities

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensIn week 12, an office ran 22% above its weather-adjusted baseline, with night load at 70% of day load. The analyst approved a question to the site manager, who found HVAC schedules had reset after a power cut. The next week's check showed use 9% above baseline, still failing, so the issue stayed open. After a second schedule fix, use was within 3% and the issue closed.

More agents for sustainability analysts

Quality control · Science and research

Scientific Unit Consistency Agent

An analysis working copy in which every calculation uses consistent, documented units.

7 steps2 checks13 jobs
Quality control · Automation

Emission Factor Update Agent

Bring every emission factor in the inventory up to date with a clear view of the effect on totals.

10 steps2 checks3 jobs
Analysis · Reporting

Target Progress Tracking Agent

Show every quarter whether each target is on track, and what would close any gap.

11 steps2 checks3 jobs
Analysis · Science and research

Life Cycle Inventory Data Gap Agent

Close or document every material data gap and show how it affects the result.

11 steps2 checks2 jobs