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 policy analysts

Policy Cost-Benefit Analysis Agent

A sourced cost-benefit comparison of policy options with sensitivity results

AnalysisGovernment
See how it worksGet access
Policy Cost-Benefit Analysis Agent: what goes in, what the agent does and what you get

What it does

When officials compare policy options, analysts must estimate costs and benefits over many years, and a single shaky assumption can decide the answer. This agent sets up the options the analyst defines, gathers cost data and published benefit estimates, and builds yearly cost and benefit streams. It calculates net present value with the agency's required discount rate. Then it varies each key assumption across its plausible range. If one change flips the ranking, it labels the result not robust, names the deciding assumption and asks the analyst for a better estimate. Every number links to its source. The analyst chooses the assumptions and the recommendation. Edge case: benefits that cannot be fairly priced, such as dignity or equity effects, are listed separately and not forced into dollars.

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, continueApprovedNo 1 STARTS WHEN Analyst defines the policy options 2 USES A TOOL Gather cost data and published benefit estimates 3 DOES Build yearly cost and benefit streams per option 4 USES A TOOL Calculate net present value with the agency discountrate 5 USES A TOOL Vary each key assumption across its plausible range 6 CHECKS THE RESULT Does the ranking hold under every single-assumptionchange? If not: label the result not robust, list the decidingassumption and ask the analyst for a better estimate.Back to step 3. 7 YOU APPROVE Analyst approves assumptions and the write-up 8 RESULT Cost-benefit table and memo
Read the steps as a list
  1. Analyst defines the policy options
  2. Gather cost data and published benefit estimates
  3. Build yearly cost and benefit streams per option
  4. Calculate net present value with the agency discount rate
  5. Vary each key assumption across its plausible range
  6. Does the ranking hold under every single-assumption change?If not: label the result not robust, list the deciding assumption and ask the analyst for a better estimate. Back to step 3.
  7. Analyst approves assumptions and the write-upThe agent waits here for your OK.
  8. Cost-benefit table and memo

How it decides

Options are ranked by net present value. If any single assumption changed within its range flips the ranking, the result is labeled not robust.

  • Use the agency's required discount rate
  • Unpriceable effects are listed, not monetized
  • Ranking flips under one assumption mean 'not robust'

Make it yours

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

  • Discount rate (default agency rate)
  • Time horizon (default 10 years)
  • Which assumptions to test
  • Output format (memo, slide, table)

What keeps you in control

It always asks you first

  • Analyst approves all assumptions and conclusions

Hard limits

  • Does not recommend a policy
  • Every figure must show its source

It stops when

  • Done: robust ranking or clear statement of what it depends on
  • Stop: no cost data for an option

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 happensFor three options to reduce youth reoffending, option B led with a $14 million NPV at a 3% rate. When the program effect size was set to the low end of the studies, 8% instead of 15%, option C led, so the robustness check failed. The agent labeled the result not robust and asked for local data. The analyst supplied a 12% estimate, B held, and she approved the memo.

More agents for policy analysts

Reporting · Research

Stakeholder Brief Update Agent

Current, documented stakeholder briefs.

8 steps1 check5 jobs
Reporting · Government

Campaign Promise Delivery Tracker Agent

A sourced status table for every promise that the official can publish with confidence

10 steps2 checks4 jobs
Creating · Automation

Amendment Comparison and Impact Agent

A short briefing note on what each amendment changes, with conflicts and knock-on effects listed and sourced

11 steps2 checks3 jobs
Analysis · Quality control

Fiscal Note Review Agent

A memo that shows whether the fiscal note's totals can be reproduced, which assumptions drive them and a range

10 steps2 checks3 jobs