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 urban planners

Variance Request Precedent Review Agent

The planner has a ranked precedent table that shows how similar past requests were decided, ready for the staff report.

ResearchCreatingGovernment
See how it worksGet access
Variance Request Precedent Review Agent: what goes in, what the agent does and what you get

What it does

Boards need consistency, but staff rarely have time to compare a new variance request with past decisions. The agent reads the request and finds past variances with similar standards, zones and conditions. It compares the findings, conditions and outcomes of each. If the match is weak, it widens the search, for example to neighboring zones or a similar hardship. It drafts a precedent table showing the case, the decision and why it is or is not alike. The planner approves it before it is added to the staff report. Edge case: a past case that was approved with conditions the new request does not meet.

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
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Variance request received 2 USES A TOOL Read the request and the standard being varied 3 USES A TOOL Search past decisions for the same standard and zone 4 DOES Rank cases by similarity 5 CHECKS THE RESULT Are there at least three strong matches? If not: Widen the search to nearby zones and similarhardships. Back to step 4. 6 DOES Compare findings, conditions and outcomes 7 DOES Note differences that might change the result 8 CHECKS THE RESULT Does each comparison cite the decision and itsreasons? If not: Reread the minutes and complete the missingreasons. Back to step 3. 9 USES A TOOL Draft the precedent table 10 YOU APPROVE Planner approves the table 11 RESULT Table added to the staff report
Read the steps as a list
  1. Variance request received
  2. Read the request and the standard being varied
  3. Search past decisions for the same standard and zone
  4. Rank cases by similarity
  5. Are there at least three strong matches?If not: Widen the search to nearby zones and similar hardships. Back to step 4.
  6. Compare findings, conditions and outcomes
  7. Note differences that might change the result
  8. Does each comparison cite the decision and its reasons?If not: Reread the minutes and complete the missing reasons. Back to step 3.
  9. Draft the precedent table
  10. Planner approves the tableThe agent waits here for your OK.
  11. Table added to the staff report

How it decides

It ranks past cases by same standard, zone type, size of relief and stated hardship, and widens the search if fewer than three good matches are found.

  • Strong match needs the same standard and zone type
  • Compare size of relief within 25%
  • Always report decisions that went the other way
  • Cite the file and date of each decision

Make it yours

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

  • Similarity criteria and weights
  • Years of decisions searched (default 10)
  • Minimum matches (default 3)
  • Table format

What keeps you in control

It always asks you first

  • Planner approves the table before it goes into the staff report

Hard limits

  • Never recommend approval or denial
  • Do not omit unfavorable precedents

It stops when

  • Done: precedent table approved
  • Stop: no similar decision exists and the planner decides how to proceed

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 happensA request for a 6 foot side yard reduction yields only 2 close matches in the zone. The agent widens to nearby zones and finds 3 more. One approved case had a condition for fencing that the new applicant did not offer. After rechecking citations, the agent drafts a table of 5 cases, 3 approved and 2 denied. The planner approves it.

More agents for urban planners

Reporting · Creating

Public Consultation Theme Evidence Agent

A consultation summary that fairly reflects every view, with traceable evidence.

7 steps1 check2 jobs
Reporting · Research

Stakeholder Brief Update Agent

Current, documented stakeholder briefs.

8 steps1 check5 jobs
Compliance · Government

Public Records Request Handling Agent

A timely, complete response with correct redactions and a record of each decision

10 steps2 checks6 jobs
Document handling · Creating

Regulatory Comment Docket Agent

A complete response table where every distinct issue in the comments has a counted entry and a drafted answer

10 steps2 checks5 jobs