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 board members

Capital Request Challenge Agent

A list of sharp questions and sensitivity results before the board votes on a spend.

Risk managementAnalysisManagementFinance
See how it worksGet access
Capital Request Challenge Agent: what goes in, what the agent does and what you get

What it does

Large spending requests arrive with thin assumptions and glowing forecasts. Board members must decide with little time to test them. This agent reads the request, extracts the assumptions, tests them against historical results and runs sensitivity, then lists questions. After answers it reruns. It decides what is weak by comparing each assumption with past actuals and a range. The member approves the final question list. After management answers, it reruns the cases and checks whether the downside still clears the board's return threshold. If not, it lists conditions the member may want attached. It never recommends how to vote. Edge case: a request assumes 30% sales growth when the past three years averaged 8%, so the agent shows the case at 8%.

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 Capital request arrives 2 USES A TOOL Read the request and extract assumptions 3 USES A TOOL Compare each assumption to historical actuals 4 DOES Run sensitivity on the key assumptions 5 DOES List the weak assumptions and the cases 6 DOES Draft questions for management 7 YOU APPROVE Member approves the final question list 8 USES A TOOL Rerun the cases with management's answers 9 CHECKS THE RESULT Does the case still meet the return threshold underthe downside? If not: add further questions and list conditions toattach. Back to step 4. 10 RESULT Question list and sensitivity summary
Read the steps as a list
  1. Capital request arrives
  2. Read the request and extract assumptions
  3. Compare each assumption to historical actuals
  4. Run sensitivity on the key assumptions
  5. List the weak assumptions and the cases
  6. Draft questions for management
  7. Member approves the final question listThe agent waits here for your OK.
  8. Rerun the cases with management's answers
  9. Does the case still meet the return threshold under the downside?If not: add further questions and list conditions to attach. Back to step 4.
  10. Question list and sensitivity summary

How it decides

It marks an assumption weak when it departs from history by more than a set range without a stated reason.

  • Flag an assumption beyond history plus 25%
  • Run a downside case at historical average
  • Check payback against the policy threshold
  • Ask for the source of each key number

Make it yours

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

  • Variance range (default 25%)
  • Downside case rule
  • Payback threshold
  • Question count
  • Data sources

What keeps you in control

It always asks you first

  • Member approves the questions sent to management

Hard limits

  • Never recommend a vote
  • Never share the model outside the board

It stops when

  • Done: questions answered and cases rerun
  • Stop: the request lacks a model to test

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 $12 million expansion assumes 30% annual growth. History averages 8%. The agent reruns at 8% and finds payback goes from 3 to 7 years. It lists four questions. Management answers with a signed contract for 40% of the volume. The rerun at 8% plus the contract gives a 4.5 year payback. The member sends the final list.

More agents for board members

Compliance · Risk management

Related Party and Conflict Disclosure Check Agent

A conflict check on every agenda item before the meeting, with recusals noted in advance.

10 steps1 check1 job
Monitoring · Management

Board Evaluation Follow-through Agent

Evaluation findings turn into owned actions that are checked for real change.

10 steps1 check2 jobs
Compliance · IT and development

Data Governance Exception Agent

Every policy exception recorded, time-limited and reviewed with controls verified

10 steps2 checks6 jobs
Document handling · Administration

Leadership Meeting Prep and Decision Log Agent

Hold every leadership meeting with complete, checked papers and a decision log that closes.

11 steps2 checks11 jobs