Model Audit - A.I. Skill
Hand Claude your underwriting model and get back an exception report on the mechanics: broken references, sum ranges that miss rows, flipped signs, hardcodes pasted over formulas, and waterfall tiers that double-pay.
This is a logic audit, not a market opinion. It will not tell you your exit cap is aggressive. It tells you whether the formulas do what you think they do, given the inputs you chose.
What It Checks
Cash flows, row by row, where most model errors actually live
Operating assumptions logic
The equity waterfall, tier by tier, for double-payment and mis-splits
Cross-sheet references for broken links, bad anchors, and hardcodes sitting where a formula should be
Structural irregularities: GP catch-up ahead of return of capital, management fee off GPR instead of EGI, acquisition fees charged at exit, replacement reserves above the line
How It Reports
Findings come back in chat, sorted Critical, Medium, Minor, and Irregularities, with exact cell references. It never edits your workbook. Zero findings is a valid result, and the skill will not manufacture issues to fill an empty section.
Perfect For
Analysts and sponsors sending models to lenders, investors, and IC. Anyone who has found a broken SUM range after the deck went out.
Requires: Claude with file uploads and code execution. Format: .skill file
Hand Claude your underwriting model and get back an exception report on the mechanics: broken references, sum ranges that miss rows, flipped signs, hardcodes pasted over formulas, and waterfall tiers that double-pay.
What It Checks
The monthly cash flow tab, row by row, where most model errors actually live
The equity waterfall, tier by tier, for double-payment and mis-splits
Cross-sheet references for broken links, bad anchors, and hardcodes sitting where a formula should be
Structural irregularities: GP catch-up ahead of return of capital, management fee off GPR instead of EGI, acquisition fees charged at exit, replacement reserves above the line
How It Reports
Findings come back in chat, sorted Critical, Medium, Minor, and Irregularities, with exact cell references. It never edits your workbook. Zero findings is a valid result, and the skill will not manufacture issues to fill an empty section.
Perfect For
Analysts and sponsors sending models to lenders, investors, and IC. Anyone who has found a broken SUM range after the deck went out.
Hand Claude your underwriting model and get back an exception report on the mechanics: broken references, sum ranges that miss rows, flipped signs, hardcodes pasted over formulas, and waterfall tiers that double-pay.
This is a logic audit, not a market opinion. It will not tell you your exit cap is aggressive. It tells you whether the formulas do what you think they do, given the inputs you chose.
What It Checks
Cash flows, row by row, where most model errors actually live
Operating assumptions logic
The equity waterfall, tier by tier, for double-payment and mis-splits
Cross-sheet references for broken links, bad anchors, and hardcodes sitting where a formula should be
Structural irregularities: GP catch-up ahead of return of capital, management fee off GPR instead of EGI, acquisition fees charged at exit, replacement reserves above the line
How It Reports
Findings come back in chat, sorted Critical, Medium, Minor, and Irregularities, with exact cell references. It never edits your workbook. Zero findings is a valid result, and the skill will not manufacture issues to fill an empty section.
Perfect For
Analysts and sponsors sending models to lenders, investors, and IC. Anyone who has found a broken SUM range after the deck went out.
Requires: Claude with file uploads and code execution. Format: .skill file
The Fractional Analyst’s
Model Audit A.I. Skill
Upgrade your underwriting process with our resources that deliver speed, accuracy, and confidence across all your deals
Hand Claude your underwriting model and get back an exception report on the mechanics: broken references, sum ranges that miss rows, flipped signs, hardcodes pasted over formulas, and waterfall tiers that double-pay.
This is a logic audit, not a market opinion. It will not tell you your exit cap is aggressive. It tells you whether the formulas do what you think they do, given the inputs you chose.
What It Checks
Cash flows, row by row, where most model errors actually live
Operating assumptions logic
The equity waterfall, tier by tier, for double-payment and mis-splits
Cross-sheet references for broken links, bad anchors, and hardcodes sitting where a formula should be
Structural irregularities: GP catch-up ahead of return of capital, management fee off GPR instead of EGI, acquisition fees charged at exit, replacement reserves above the line
How It Reports
Findings come back in chat, sorted Critical, Medium, Minor, and Irregularities, with exact cell references. It never edits your workbook. Zero findings is a valid result, and the skill will not manufacture issues to fill an empty section.
Perfect For
Analysts and sponsors sending models to lenders, investors, and IC. Anyone who has found a broken SUM range after the deck went out.
Requires: Claude with file uploads and code execution. Format: .skill file