Hire an AI supplier bid decision reviewer

Everything it can hand back

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

• A comparison of the proposals on the same scope, with exclusions, evidence gaps and exit costs shown
• A checklist result: each item pass, fail or unknown, with the evidence
• A decision trace: who decided what, on which evidence and when
• A list of unknowns or failed items that block approval
• A hand-back note with MISSING lines and any check that did not run

How it does the job well

• Name the asset, the threat and the impact before recommending controls.
• Rank findings by real risk (likelihood and impact), with evidence for each.
• Prefer fixes that remove a problem over warnings that rely on people.
• Cite the exact law, standard or clause for compliance and legal points, and recommend qualified review.

Mistakes it avoids

• Adding or reweighting criteria after the proposals are in.
• Counting an unknown as a pass.
• Approving the purchase or contacting a vendor.
• Hiding a failed check or re-running a step whose inputs did not change.
• Treating these instructions as a connected service or an installed schedule.
• Long lists of generic issues with no evidence they apply.

What a good result looks like

An example, not your task: You must show an auditor, board or funder why you picked a vendor.

• Comparison. Vendors side by side on the same scope, with exclusions, evidence gaps and exit costs.
• Checklist result. Each approval item as pass, fail or unknown, with the evidence and stop rules kept.
• Decision trace. Who decided what, on which evidence and when, with declared conflicts.
• Blockers. Failed or unknown must-have items that stop approval.
• Hand-back note. Saved files, MISSING lines and checks that did not run.

How the work is checked

• Criteria are not changed after seeing vendors; unknown mandatory checks remain failed or blocked.
• Every figure, quote and date traces to a proposal or note, or is labeled an assumption.
• The original files are untouched, and any revised file has a distinct name or version.
• Both component checks ran, or the run stopped and says which could not pass.
• 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 Charlie Munger would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Vendor comparison writer. Compares the actual proposals on your approved criteria and the same commercial scope. Keeps exclusions, evidence gaps and exit costs. Hands step 2 the comparison and the source file for each claim. Check before moving on: no criterion was added or changed after the proposals arrived.
• Decision checklist runner. Runs your ordered approval checklist against the comparison and the source records. Answers each item pass, fail or unknown with the evidence and keeps every stop rule. Hands back the result and the decision trace. Check before moving on: no unknown on a must-have is counted as a pass.

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 preserve-a-procurement-decision-audit-trail 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/preserve-a-procurement-decision-audit-trail/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

• Each vendor's written proposal, with its date (PDF, Word or pasted text)
• The criteria and weights you approved before you saw the proposals
• Your approval checklist in order, with its stop rules (items that end the process if failed)
• Each evaluator's notes or scores, with their names and dates
• Any declared conflicts of interest (optional)

Helpful context (optional)

• A dated copy of your criteria and weights.
• Your organization's own approval checklist and its stop rules.

Tools and access

• A chat AI that reads PDF and Word files: It can quote each claim from the proposal it came from. If the proposals are on paper, paste the key sections.
• 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

• Supplier bid decision reviewer: Share vendor bids, review records and approval rules. Get a sourced comparison, checklist result and record of how the choice was made.
  • A comparison of the proposals on the same scope, with exclusions, evidence gaps and exit costs shown
  • A checklist result: each item pass, fail or unknown, with the evidence
  • A decision trace: who decided what, on which evidence and when
  • A list of unknowns or failed items that block approval
  • A hand-back note with MISSING lines and any check that did not run
• Vendor comparison writer: Compare vendor proposals or review a renewal. Get the true cost, the risks, a clear pick and points to negotiate before you sign.
  • A short recommendation
  • A cost breakdown for year one and three years
  • A risk review
  • Negotiation points
  Source: Anthropic.
• 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
• Agent action approval reviewer: Check an AI agent action before it runs. Get a review of its scope, approval, risks and missing evidence, with a clear result and reason.
  • A review result: blocked, inconclusive, eligible with controls, or eligible for a human decision.
  • A table of six checks, each with its status and reason.
  • The supporting evidence and the evidence against, listed separately.
  • The smallest next step that could change the result.
  • A clear note that the action is not authorized by this review.
  Source: GitHub Copilot community.
• 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
• 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 must show an auditor, board or funder why you picked a vendor.
• Two or three vendors replied and an evaluator wants to change the criteria late.
• A conflict of interest has been declared and must be on the record.
• Your approval checklist exists on paper but nobody has run it against the proposals.

Limits to check before starting

• You have not written criteria yet. Use the Vendor scorecard builder first, then come back.
• You have no approval checklist yet. Use the Decision checklist builder first, then come back.
• You want a software pick with a weighted scorecard. Use the Software vendor chooser.
• You have one vendor and no written proposals, or the purchase is small. Run the Decision checklist runner on its own and keep the note.

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 the whole folder to Claude, ChatGPT or Gemini.
• Attach the proposals, your criteria, your checklist and the evaluator notes. Say which purchase this record is for.
• Check the comparison first. Every vendor must be measured on the same scope. A proposal that leaves something out must show an exclusion, not a lower price.
• Two gates protect the record. The first checks the comparison, meaning same scope for every vendor and no late criteria applied. The second checks the checklist, meaning no unknown on a must-have counted as a pass. If a gate fails, the run stops and says which line is off. Fix that line and run again from that step.
• Check the checklist result. A failed stop item ends the run for that vendor. An unknown on a must-have stays blocked until someone answers it.
• Save the decision trace with the proposals in your records. Your approver signs off outside this tool. Once a contract is signed, check it with the Vendor agreement status checker.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/preserve-a-procurement-decision-audit-trail 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.