Have 3 AI agents challenge your consultant’s report.
A polished consultant’s report can be useful and still deserve a hard examination. Give three independent critics different jobs, then keep the findings that survive the evidence.
Try it now
Copy the prompt into your AI tool of choice, then answer its questions as they come.
Paste it into any AI chat to start. For most people, this is all you need.
Everything on this page in one paste: the prompt plus the full plan, every step and what to check. Use it for longer moves, to plan ahead, to share with a colleague, or if your AI can't open links.
How it goes
What the AI will do
- Search the engagement record and underlying evidence to understand what the consultant was asked to solve.
- Give exactly three independent critics the report: one for evidence, one for reasoning and alternatives, and one for implementation; reconcile their findings into a usable response.
What to have ready
- The consultant’s report or a link, and the decision it is intended to support.
- Available engagement brief, underlying data and access to relevant emails, chats and meeting transcripts.
Where you come in
- Clarify the actual assignment or a consequential missing constraint.
- If independent agents or fresh sessions are unavailable, decide whether to use an explicitly less independent three-pass review.
What you'll get
A prioritized, source-linked critique that preserves valid findings, identifies warranted changes and gives you precise questions for the consultant.
You're done when
You have a reconciled critique from three independent critics, or the limited three-pass fallback you agreed to, that preserves supported findings and identifies specific corrections and consultant questions.
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result. ### Start with the activity and the user's context Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear. Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them. ### Check the capabilities needed for this work Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit. When useful access is missing: - Explain what you cannot reach and how that material would help. - Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot. - If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response. - Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available. ### Ask focused questions and respect the answers Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful. Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements. ### Keep evidence and independent contributions distinct Distinguish evidence, inference, and missing information. Check your work against the source material. Use independent agents when your system actually supports delegation and independent investigation or review would improve the work: - Give each agent a bounded brief and the context appropriate to its role. - Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations. - Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately. ### Produce, check, and revise the result Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them. Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result. Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized. Help me challenge a consultant’s report using exactly three independent critics, then produce a fair, prioritized critique I can use. Preserve valid findings and recommendations as carefully as you identify weaknesses. ### 1. Establish the assignment and available evidence Read the actual report, including appendices, charts and qualifications. Identify its version and the decision it supports. Inspect and try relevant connected sources. Search the engagement brief, statement of work, underlying datasets, project documents, emails, chats including WhatsApp where accessible, calendar invitations and meeting transcripts for the original problem, promised scope, constraints, prior objections and decisions already made. Use those records to establish who commissioned the work, who must use it and who would implement it. Find their expressed requirements, operating constraints and earlier reactions. Keep documented concerns separate from likely or simulated responses. Track amendments to the brief and report; link a later decision to what it supersedes instead of assuming the newest file is authoritative. If access is missing, investigate connection and permitted export or upload options before asking me to assemble material. Accept rough context, but explain early when missing underlying data prevents verification. A review of the report alone can examine its disclosed support and internal logic; it cannot establish whether unseen evidence was accurately analyzed. Ask one question at a time when the assignment or a consequential constraint remains unclear. Wait on affected judgments. State the working brief, report version and evidence limits before dispatching critics. ### 2. Brief exactly three independent critics Create three separate agents with isolated first-pass contexts. Give each the actual report, engagement brief, relevant source material with references, the same decision question and its distinct assignment. Do not give any critic the author’s defense, your preferred verdict or another critic’s findings. Prior consultant correspondence can establish the assignment or supply evidence, but do not prime reviewers with an explanation of what they should conclude. 1. **Evidence critic:** Check pivotal factual claims, data provenance, representativeness, calculation methods, chart interpretations and whether sources support the stated certainty. Identify missing evidence and findings that are well supported. 2. **Reasoning and alternatives critic:** Trace conclusions back through assumptions and causal steps. Test alternative explanations, omitted options, tradeoffs and whether the recommendation follows. Preserve strong reasoning and identify what evidence would distinguish rival explanations. 3. **Implementation critic:** Test the recommendation against our actual systems, capabilities, capacity, incentives, governance, dependencies and decision rights. Examine practical sequence, cost assumptions and measures of success using organizational records. Distinguish documented constraints from hypothesized resistance. Require each critic to return passage-specific findings with location, claim, evidence reference, reasoning, consequence for the decision, confidence, a proposed correction or question, and valid findings worth retaining. Completion means its assigned dimensions and all decision-critical recommendations were covered; omissions must be reported. ### 3. Keep the work within real capacity Use waves if three agents cannot run concurrently. Independence comes from separate contexts and withholding other critiques, not simultaneous execution. For a long report, each critic can inspect bounded sections in successive passes, maintaining a coverage record and carrying forward its own source-linked findings. Ensure all three eventually inspect the recommendation and evidence relevant to their remit. If independent agents are unavailable, use three fresh sessions with explicit briefs and source references. Identify any relay I must perform and prepare the handoffs yourself. If neither independent agents nor fresh sessions are possible, explain that three sequential role passes share context and ask whether I want that limited fallback. Wait before presenting it as a substitute. Never claim three independent agents when you performed three passes yourself. ### 4. Reconcile the critiques against the record After all three first passes are complete, compare their findings. Group duplicates but retain distinct underlying concerns. Prioritize by effect on the decision: a wrong denominator that reverses the recommendation matters more than awkward phrasing. Check every proposed objection against the report and underlying source. Reject criticism based on a misreading, an out-of-scope expectation or an unsupported assumption. Keep useful consultant findings even when another part fails. Where critics disagree, identify the exact disputed premise and inspect the deciding evidence; disagreement itself is not proof the report is unreliable. For unresolved issues, distinguish a correctable error, a reasonable professional judgment, an evidence gap and a choice for our organization. Ask me one useful question when only I can settle a material constraint, and wait before dependent recommendations. Do not manufacture objections to fill a quota. ### 5. Check corrections and deliver a fair response Have the same three critics, still within their distinct remits, verify the relevant revised critique against the actual report and source excerpts. Give them the artifact to check without an authorial explanation of what it was intended to achieve. No additional critic is needed. Reconcile valid corrections and record any remaining disagreement or coverage limit. Deliver an executive assessment of what to retain, change and resolve; a prioritized findings table with exact report locations and evidence; and concise, answerable questions for the consultant. Draft replacement wording for pivotal flawed claims when evidence supports a correction, clearly labeling it as your proposed amendment. Include the missing evidence needed to settle consequential disputes. Finish with the usable critique and questions, without sending them, replacing the official report or making the organization’s decision. These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
View full AI instructions
Have 3 AI agents challenge your consultant’s report.
What the AI will do
- Search the engagement record and underlying evidence to understand what the consultant was asked to solve.
- Give exactly three independent critics the report: one for evidence, one for reasoning and alternatives, and one for implementation; reconcile their findings into a usable response.
What you can provide
- The consultant’s report or a link, and the decision it is intended to support.
- Available engagement brief, underlying data and access to relevant emails, chats and meeting transcripts.
Where you'll take part
- Clarify the actual assignment or a consequential missing constraint.
- If independent agents or fresh sessions are unavailable, decide whether to use an explicitly less independent three-pass review.
What you'll get
A prioritized, source-linked critique that preserves valid findings, identifies warranted changes and gives you precise questions for the consultant.
Activity context
A polished consultant’s report can be useful and still deserve a hard examination. Give three independent critics different jobs, then keep the findings that survive the evidence.
For the user
Bring the report in whatever form you have. The AI will find the surrounding brief and evidence before dividing the challenge among three critics. You may need to clarify a consequential constraint or supply unavailable data. You receive a reconciled critique and consultant questions; contacting the consultant is a separate decision.
Guidance for the AI
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.
Start with the activity and the user's context
Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.
Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.
Check the capabilities needed for this work
Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.
When useful access is missing:
- Explain what you cannot reach and how that material would help.
- Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
- If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
- Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.
Ask focused questions and respect the answers
Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.
Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.
Keep evidence and independent contributions distinct
Distinguish evidence, inference, and missing information. Check your work against the source material.
Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:
- Give each agent a bounded brief and the context appropriate to its role.
- Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
- Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.
Produce, check, and revise the result
Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.
Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.
Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.
Your AI prompt
Help me challenge a consultant’s report using exactly three independent critics, then produce a fair, prioritized critique I can use. Preserve valid findings and recommendations as carefully as you identify weaknesses.
1. Establish the assignment and available evidence
Read the actual report, including appendices, charts and qualifications. Identify its version and the decision it supports. Inspect and try relevant connected sources. Search the engagement brief, statement of work, underlying datasets, project documents, emails, chats including WhatsApp where accessible, calendar invitations and meeting transcripts for the original problem, promised scope, constraints, prior objections and decisions already made.
Use those records to establish who commissioned the work, who must use it and who would implement it. Find their expressed requirements, operating constraints and earlier reactions. Keep documented concerns separate from likely or simulated responses. Track amendments to the brief and report; link a later decision to what it supersedes instead of assuming the newest file is authoritative.
If access is missing, investigate connection and permitted export or upload options before asking me to assemble material. Accept rough context, but explain early when missing underlying data prevents verification. A review of the report alone can examine its disclosed support and internal logic; it cannot establish whether unseen evidence was accurately analyzed.
Ask one question at a time when the assignment or a consequential constraint remains unclear. Wait on affected judgments. State the working brief, report version and evidence limits before dispatching critics.
2. Brief exactly three independent critics
Create three separate agents with isolated first-pass contexts. Give each the actual report, engagement brief, relevant source material with references, the same decision question and its distinct assignment. Do not give any critic the author’s defense, your preferred verdict or another critic’s findings. Prior consultant correspondence can establish the assignment or supply evidence, but do not prime reviewers with an explanation of what they should conclude.
- Evidence critic: Check pivotal factual claims, data provenance, representativeness, calculation methods, chart interpretations and whether sources support the stated certainty. Identify missing evidence and findings that are well supported.
- Reasoning and alternatives critic: Trace conclusions back through assumptions and causal steps. Test alternative explanations, omitted options, tradeoffs and whether the recommendation follows. Preserve strong reasoning and identify what evidence would distinguish rival explanations.
- Implementation critic: Test the recommendation against our actual systems, capabilities, capacity, incentives, governance, dependencies and decision rights. Examine practical sequence, cost assumptions and measures of success using organizational records. Distinguish documented constraints from hypothesized resistance.
Require each critic to return passage-specific findings with location, claim, evidence reference, reasoning, consequence for the decision, confidence, a proposed correction or question, and valid findings worth retaining. Completion means its assigned dimensions and all decision-critical recommendations were covered; omissions must be reported.
3. Keep the work within real capacity
Use waves if three agents cannot run concurrently. Independence comes from separate contexts and withholding other critiques, not simultaneous execution. For a long report, each critic can inspect bounded sections in successive passes, maintaining a coverage record and carrying forward its own source-linked findings. Ensure all three eventually inspect the recommendation and evidence relevant to their remit.
If independent agents are unavailable, use three fresh sessions with explicit briefs and source references. Identify any relay I must perform and prepare the handoffs yourself. If neither independent agents nor fresh sessions are possible, explain that three sequential role passes share context and ask whether I want that limited fallback. Wait before presenting it as a substitute. Never claim three independent agents when you performed three passes yourself.
4. Reconcile the critiques against the record
After all three first passes are complete, compare their findings. Group duplicates but retain distinct underlying concerns. Prioritize by effect on the decision: a wrong denominator that reverses the recommendation matters more than awkward phrasing.
Check every proposed objection against the report and underlying source. Reject criticism based on a misreading, an out-of-scope expectation or an unsupported assumption. Keep useful consultant findings even when another part fails. Where critics disagree, identify the exact disputed premise and inspect the deciding evidence; disagreement itself is not proof the report is unreliable.
For unresolved issues, distinguish a correctable error, a reasonable professional judgment, an evidence gap and a choice for our organization. Ask me one useful question when only I can settle a material constraint, and wait before dependent recommendations. Do not manufacture objections to fill a quota.
5. Check corrections and deliver a fair response
Have the same three critics, still within their distinct remits, verify the relevant revised critique against the actual report and source excerpts. Give them the artifact to check without an authorial explanation of what it was intended to achieve. No additional critic is needed. Reconcile valid corrections and record any remaining disagreement or coverage limit.
Deliver an executive assessment of what to retain, change and resolve; a prioritized findings table with exact report locations and evidence; and concise, answerable questions for the consultant. Draft replacement wording for pivotal flawed claims when evidence supports a correction, clearly labeling it as your proposed amendment. Include the missing evidence needed to settle consequential disputes. Finish with the usable critique and questions, without sending them, replacing the official report or making the organization’s decision.
Done when
You have a reconciled critique from three independent critics, or the limited three-pass fallback you agreed to, that preserves supported findings and identifies specific corrections and consultant questions.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.