Hire an AI pilot decision checker

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A pilot evidence brief on what you tested, against what baseline, and what the numbers show
• A list of data gaps, such as no baseline, a small sample or failed cases left out
• Your checklist run item by item, with evidence and a pass, fail or unknown mark
• A go, wait or stop call with the reason, and the one fact that would change it
• A hand-back summary with every gap named

How it does the job well

• Never write or change a success measure to fit the results. If the goal was written after the results, say so.
• A demo on chosen examples is not a test on typical work. Say which one the user has.
• A missing baseline, privacy answer or mandatory checklist item blocks the decision it touches. Write MISSING: and carry on with the rest.
• Keep measured results apart from stories and opinions.
• Do not invent figures, costs, case counts or a checklist. If the user has no checklist, say so and offer to help write one separately.
• Clarify the purpose, audience and limits, such as length, format and deadline, first.
• Use the user's materials and voice. Ask for samples when tone matters.

Mistakes it avoids

• Generic output that ignores the user's details.
• Inventing facts, quotes or sources.
• Long answers when a short one was asked for.

What a good result looks like

An example, not your task: You ran an AI pilot for a few weeks and need to decide whether to roll it out.

• Pilot evidence brief. The goal, the baseline, the signals watched and what the numbers show, with sources named.
• Data gaps. Missing baselines, small samples or left-out failures, each with what it affects and how to close it.
• Checklist result. Each item with its evidence and a pass, fail or unknown mark, and the stop rule result.
• Call. Go, wait or stop, the reason in one line, and the one fact that would change it.
• Hand-back summary. Each saved file, every MISSING line and the checks that did not run.

How the work is checked

• The goal and the baseline are stated and dated, or marked MISSING.
• Every percentage shows its denominator.
• Failed cases appear in the evidence.
• Every checklist item has evidence and a mark, and unknowns are counted as fails.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Fei-Fei Li and Charlie Munger would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Hypothesis-first design planner. Restates the pilot as something you could prove wrong, with the signals you said you would watch and the baseline. It does not rewrite the goal to fit the results. Hands step 2 the goal and the signals. Check before moving on: the goal and the baseline are written, or marked MISSING.
• Spreadsheet checker. Reads your case table. Checks how many cases, how many failed, costs and totals. Separates measured results from stories. Hands step 3 a list of checked figures and data gaps.
• Decision checklist runner. Runs your go or stop checklist in order against the evidence. Marks each item pass, fail or unknown. An unknown counts as a fail. A failed stop item means stop. Makes no change to any budget or rollout.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id prepare-an-ai-pilot-go-decision only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/prepare-an-ai-pilot-go-decision/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• The pilot goal in one or two sentences, written before the pilot if you have it
• The baseline: how the work was done and measured before AI, with dates
• The test cases and results, as a sheet or a table, with failures included
• What the pilot cost and what rollout would cost (optional)
• Your go or stop checklist, with the pass line for each item

Helpful context (optional)

• The written goal from before the pilot started.
• A named person who owns the go or stop call.

Tools and access

• Claude, ChatGPT or Gemini with file support: They can open your results sheet and check the totals. In plain chat, paste the table as text.
• A spreadsheet app such as Excel or Google Sheets (optional): Lets you recalculate the totals yourself and compare them with the AI's.
• Any good chat AI: This is text reasoning that any strong chat AI handles with the skill.
• A notes app or document: Keeps the log, the checklist and your records in one place you can come back to.

Functions and deliverables

• AI pilot decision checker: Share your pilot goals, baseline, real results and approval rules. Get checked findings, data gaps and a go-or-stop checklist result.
  • A pilot evidence brief on what you tested, against what baseline, and what the numbers show
  • A list of data gaps, such as no baseline, a small sample or failed cases left out
  • Your checklist run item by item, with evidence and a pass, fail or unknown mark
  • A go, wait or stop call with the reason, and the one fact that would change it
  • A hand-back summary with every gap named
• Lean UX experiment planner: Describe a design problem. Get assumptions to check and small tests to plan before you build.
  • A ranked assumptions list
  • Hypotheses with success rules
  • Lightweight experiment plans
  • A score of your current process with fixes
  Source: Wondel.ai.
• Spreadsheet checker: Check a spreadsheet for odd totals, broken formulas and messy data. Your original stays safe, and you get findings with proof and a new file.
  • A sheet inventory and scope
  • Findings with evidence locations
  • Data quality and formula issues
  • A new output file with your edits
  • Reconciliation results
  Source: Burhan Sebin.
• Decision checklist runner: Run your checklist on a real decision. Each item gets a fact, a pass, fail or unknown, then a clear go, wait or no call.
  • Each item answered with evidence, marked pass, fail or unknown
  • Which stop items failed, if any
  • The call: go, wait or no, with a reason
  • A record of what you expect and what would show you wrong
• me.md builder: Answer a few questions, or let your AI use what it remembers. Get one private file that tells any AI who you are.
  • One me.md file with up to ten dated sections
  • A list of facts taken from memory, for you to check
  • A short note of what it removed for privacy
  • The two next steps, to save it and set it up
• Fei-Fei Li: Review an AI plan, dataset or test using Fei-Fei Li’s public approach. Start with affected people, check the examples and keep human judgement involved.
  • Your problem worked on the way Fei-Fei Li would: their questions, their method, their push-back
  • The answer or the work, in Fei-Fei Li's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Charlie Munger: Stress-test a plan using Charlie Munger’s recorded approach. Turn it upside down, examine incentives and ask where your knowledge ends.
  • Your problem worked on the way Charlie Munger would: their questions, their method, their push-back
  • The answer or the work, in Charlie Munger's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• You ran an AI pilot for a few weeks and need to decide whether to roll it out.
• A vendor or a colleague says the pilot worked and you want to see the numbers behind it.
• You need a one-page evidence brief for a manager, a board or a budget holder.
• Your pilot has no baseline or no written checklist and you want to know what is missing before you decide.

Limits to check before starting

• You have not started the pilot. Use the AI build-or-buy advisor to decide what to try first.
• You are checking one AI agent before it goes live. Use the AI agent launch security reviewer workflow.
• You only need to check one results sheet. Use the Spreadsheet checker.
• You want the decision made for you. This workflow prepares the evidence. A person decides.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Download the pack and give it to your AI, such as Claude, ChatGPT or Gemini. Use one that can open spreadsheets if your results are in a sheet.
• Attach the results sheet. Paste the goal, the baseline, the costs and your checklist.
• After step 1, check that the goal was written before the results. Say so if it was changed afterwards.
• After step 2, check the case count and the failed cases against your own sheet. Fix any number that is wrong.
• After step 3, read each unknown. Find the fact, then run the checklist again. A person makes the final call.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/prepare-an-ai-pilot-go-decision for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.