For anyone who signs off on a pack that goes to a buyer, an IC, or a board: six agent desks that run the sell side pack end to end, plus the four stop conditions that halt a run before it ships, prose versus schedule, cross page totals, stale source, and unsourced claim.
A short walkthrough of the exact mechanic: where the prompts go, what to answer when the first one asks, and what a good first output looks like. Same for every pack in the library.
Never used Claude? It is free and takes 30 seconds to open. Copy the instruction below, paste it into Claude, and it reads this page and walks you through everything, one question at a time.
One tap. It changes how much Claude explains, and how slowly it goes. You can change it any time.
A short instruction plus a link to this page lands on your clipboard. First copy asks for your email once. That unlocks every button across the whole library.
Free account, no card, 30 seconds. This tab stays open so you can come back.
Open claude.ai ↗Claude reads this page, asks one question about your work, then guides you step by step until your first output is right. If anything looks wrong, tell Claude what you see, and it fixes it with you.
I am comfortable copy-pasting and following instructions, but I am not a developer. There is nothing to install for this one and no commands to type: it all happens inside Claude. If any instruction below implies a Terminal, translate it into the equivalent click path for me instead. - Plain English. Define jargon the first time it appears. - One step at a time, then wait for me to confirm before the next one. - Tell me what success looks like at each step, and diagnose any error before moving on. Follow the instructions below with those rules applied. This page is a MAP, not a vault. There are no prompts on it. It lists six desks that each live on their own page, plus the four stop conditions that apply across all of them. Your job is to help me pick ONE desk and send me to that desk's page, where the prompts actually live. Do not tell me the prompts are here. The six destinations: - Model Desk, the linked income statement, balance sheet, and cash flow tied out to filings. https://consultance.ai/library/three-statement-model-vault - Comps Desk, peer screening, fiscal year calendarization, and outlier flags. https://consultance.ai/library/trading-comps-vault - Teaser Desk, the anonymised two page teaser with an anonymization gate before assembly. https://consultance.ai/library/sell-side-teaser-kit - Sector Desk, the sector map from public filings, market size, players, margin pools. https://consultance.ai/library/sector-overview-engine - Buyer List Desk, the buyer and investor list alongside the data room. https://consultance.ai/library/deal-room-autopilot - Reconciliation Desk, checks every figure across every page and blocks release while two disagree. https://consultance.ai/library/ai-output-verification-pack Routing rules: - Ask exactly ONE question in your first message, nothing else. No preamble. - After my answer, name ONE desk, give the exact URL, and say in one sentence why it fits. Offer a second desk only if I ask. - If I say I am starting a sell side pack from scratch, route me to the Model Desk first, because everything downstream reads off it. - If I say the pack is already drafted and I am worried about errors, route me to the Reconciliation Desk. - Tell me to open that page and use its own guided setup there. The install steps and the prompts belong to that page, not this one. - Pin Claude Opus 5 for the model, teaser, and reconciliation work. - Do not invent prompts, commands, or file names to fill the gap. If I ask for a prompt, send me to the destination page. - The four tripwires on this page catch arithmetic and provenance failures, not a wrong judgement that is internally consistent. A named MD, partner, or controller signs the pack before it reaches a buyer, an IC, or a board. First message: ask me "Which seat are you in on this deal, and what are you building right now: the model, the comps, the teaser, the sector overview, the buyer list, or a check on a pack that is already drafted?" Then route me to one desk.
Got the prompts. Want them wired into your actual stack? We map that on a free AI audit.
Book the free auditFor anyone who signs off on a pack that goes to a buyer, an IC, or a board: six agent desks that run the sell side pack end to end, plus the four stop conditions that halt a run before it ships, prose versus schedule, cross page totals, stale source, and unsourced claim.
The setup rail and every prompt above are free and stay free. The cost is your time, and the risk of wiring it wrong on live data.
Back to the prompts ↑The desks are six separate guides and the tripwires are yours to maintain. Consultance stands the assurance layer up in your own environment: the stop conditions wired to your house standard, your VDR and model library connected under governed access, and a release gate your reviewers actually sign. Reply "wire it" for a 30-minute slot.
Book a build call →Prompt sets authored by consultance.ai. Each desk carries its own honest ceiling, stated on its own page. Tripwires catch arithmetic and provenance failures, they do not catch a wrong judgement that is internally consistent. This is decision support, not investment, legal, or accounting advice. A named MD, partner, or controller signs the pack before it reaches a buyer, an IC, or a board. Your data stays in your own Claude tenant, we never see it.
Want this wired into your stack instead of running it yourself? That is our AI deal desk and finance automation service.
The AI tools, launches, and shifts that actually matter, in plain English. New library drops the moment they land.
What is in this stack
For CFOs and PE deal teams: 12 prompts that build a tied-out income statement, balance sheet, and cash flow from your own filings inside Claude Opus 5. A prompt vault that replaces a first-year analyst grind and Wall Street Prep templates.
For PE deal teams & equity research analysts who rebuild trading comps by hand every week: 12 prompts that screen the peer set, calendarize the fiscal years, reconcile every multiple a second way, & land one MD ready sheet in Claude. Kills the weekly manual rebuild.
For PE deal teams and sell-side principals: 11 prompts that turn raw financials into the anonymized 2-page M&A teaser a banker bills a retainer to write, run inside your own Claude. No CIM, no analyst all-nighter, and nothing leaves your machine.
For PE deal teams and corp dev: 10 Claude prompts that build a full sector overview from public filings, market size, players, margin pools, the shift incumbents miss. Replaces the McKinsey market map engagement.
For PE deal teams and corp dev: read a data room and draft the IC memo in Claude with 11 prompts. Reads every contract, ties the financials to source, flags concentration and change of control, and reconciles before the memo. Replaces the associate grind and a chunk of the third party diligence engagement.
For anyone who signs off on an AI-assisted report, model, or board pack: 13 Claude prompts that fact-check every citation, quote, and number before you ship it.
The person drowning in this every week is the one who'll actually want it.
The Agent Tripwire Pack is a finance and data build in the consultance.ai AI Build Library. For anyone who signs off on a pack that goes to a buyer, an IC, or a board: six agent desks that run the sell side pack end to end, plus the four stop conditions that halt a run before it ships, prose versus schedule, cross page totals, stale source, and unsourced claim. It fits PE deal teams and principals, fund controllers, CFOs, and family office principals who have watched a figure in the write up disagree with the schedule, survive two rounds of human review, and get caught by the buyer on the call. Setup difficulty is Medium, with 4 plain-English steps.
For anyone who signs off on a pack that goes to a buyer, an IC, or a board: six agent desks that run the sell side pack end to end, plus the four stop conditions that halt a run before it ships, prose versus schedule, cross page totals, stale source, and unsourced claim.
It fits PE deal teams and principals, fund controllers, CFOs, and family office principals who have watched a figure in the write up disagree with the schedule, survive two rounds of human review, and get caught by the buyer on the call.
Medium to set up — one guided setup instruction covering 4 plain-English steps on the resource page.
The desks are six separate guides and the tripwires are yours to maintain. Consultance stands the assurance layer up in your own environment: the stop conditions wired to your house standard, your VDR and model library connected under governed access, and a release gate your reviewers actually sign. Reply "wire it" for a 30-minute slot.
Prompt sets authored by consultance.ai. Each desk carries its own honest ceiling, stated on its own page. Tripwires catch arithmetic and provenance failures, they do not catch a wrong judgement that is internally consistent. This is decision support, not investment, legal, or accounting advice. A named MD, partner, or controller signs the pack before it reaches a buyer, an IC, or a board. Your data stays in your own Claude tenant, we never see it.
The Agent Tripwire Pack is the starting point. On a free AI audit we map where it fits your stack and what consultance.ai would build around it.
This build comes from our AI consulting and AI implementation practice — see the full AI in finance guide and how we work with CFO teams.