Stats
Actions
Tags
From deal-diligence
**SaaS / Operating Model Financial Template Builder**: Reads a source Excel operating model (P&L, Balance Sheet) and generates a standardized 3-tab financial model with Input Page, Financial Model Template, and Output Tab. - MANDATORY TRIGGERS: "build financial model", "run the template", "create financial model template", "3-tab model", "generate financial model", "financial model from P&L" - Use this skill whenever a user uploads an Excel file containing a P&L or income statement and wants a structured financial model built from it - Also triggers for: "Input Page + FMT + Output Tab", "6+6 analysis", "go-forward analysis", "SaaS model template", "turn this P&L into a model", "build a model from this", "model this out", "create a financial template", "build the pattern template model"
How this skill is triggered — by the user, by Claude, or both
Slash command
/deal-diligence:financial-model-builderThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill takes a user's source Excel operating model (containing a P&L/Income Statement and
This skill takes a user's source Excel operating model (containing a P&L/Income Statement and optionally a Balance Sheet) and builds a standardized 3-tab financial model on top of it:
The skill is adaptive: it detects whether the source data is monthly, quarterly, or annual, adapts the column structure and formulas accordingly, and handles any number of cost categories.
Read the full blueprint before writing any code:
Read: {SKILL_DIR}/references/blueprint.md
To find {SKILL_DIR}: look at the path you used to read this SKILL.md — it ends in
.../financial-model-builder/SKILL.md. The {SKILL_DIR} is that directory.
The blueprint is the complete row-by-row specification for all 3 tabs, including every formula, formatting rule, border pattern, number format, and cross-tab dependency. Read it in full before writing any code — it contains critical details that are easy to get wrong.
Also read the xlsx skill reference before building:
Read: {SKILL_DIR}/../xlsx/SKILL.md
Before building, collect these inputs (ask if not provided):
| Parameter | Example | Notes |
|---|---|---|
| Source P&L tab name | "P&L", "Income Statement" | Which tab has the P&L data |
| Last actuals period | "Dec 2024", "April 2025" | Last period with actual (not forecast) data |
| Forecast period | "2 years", "3 years" | How many years of forecast columns |
| Company name | "Acme Corp" | Used in titles and headers |
| Balance Sheet tab | "Balance Sheet", or "none" | Optional — skip BS/CF if not available |
Write a Python script to scan the source file and detect:
1. Periodicity — are columns monthly, quarterly, or annual?
2. Column range — where does data start/end? Which row has period headers?
3. Row structure — scan for key line items in column B/D:
4. Units — actual dollars, thousands, or millions? Check revenue magnitude.
5. Category detection — group expense lines into logical categories. The reference template uses S&M, R&D, G&A but adapt to whatever is in the source (e.g., "Direct Costs" / "Indirect Costs", or 10+ individual departments).
Follow the blueprint (section "TAB 1: INPUT PAGE") exactly. Key rules:
#4472C4showGridLines = False)#0000FF), thin blue top border#DDDDDD), thin top+bottom bordersAdaptive category slots: The reference has 5 categories with 8 subsegment slots each (Revenue, COGS, S&M, R&D, G&A). If the source has more/fewer categories:
Follow the blueprint (section "TAB 2: FINANCIAL MODEL TEMPLATE") exactly. Critical rules:
#548235#FFFFFF) on ALL data cellsColumn structure (adaptive):
Formula pattern for data cells:
=IF($D{n}="","",IFERROR(INDEX('{Source}'!$C${start}:$BD${end},
MATCH($D{n},'{Source}'!$B${start}:$B${end},0),
MATCH(H$1,'{Source}'!$C$4:$BD$4,0))/$B$1,0))
This 2D INDEX/MATCH finds the subsegment by name in the source P&L, then finds the period column by matching the date in row 1. Adapt match row/range based on Step 2 findings.
Totals use SUM formulas (not source references): =SUM(H14:H21) for Total Revenue, etc.
Percentage rows: =IF(H$22=0,0,H34/H$22) — always guard against division by zero.
YoY %: Start 4 periods in (quarterly) or 12 periods in (monthly):
=IFERROR(H22/D22-1,"")
Follow the blueprint (section "TAB 3: OUTPUT TAB") exactly. Most formatting-intensive tab.
#BF8F00Column layout (always this structure regardless of source periodicity):
| Col | Content | Notes |
|---|---|---|
| E–H | Annual actuals (CY years) | |
| I | Spacer | Hidden |
| J–K | N-months actual vs. N-months prior year | 6+6 split |
| L–M | $ and % variance | |
| N | Spacer | Hidden |
| O–P | "To Go" months actual vs. prior year | |
| Q–R | $ and % variance | |
| S | Spacer | Hidden |
| T | LTM (last twelve months) | |
| U | Spacer | Hidden |
| V–X | Budget / Forecast years |
SUMIFS formula pattern (ALL data cells):
=SUMIFS('Financial Model Template'!$H{row}:${lastcol}{row},
'Financial Model Template'!$H$3:${lastcol}$3,">="&{col}$2,
'Financial Model Template'!$H$3:${lastcol}$3,"<="&{col}$3)
Row structure must include ALL segments — not just summary totals. Every subsegment, category total, gross profit, EBITDA line, and cash flow item needs data formulas.
Variance columns:
=K{n}-J{n} (or P-O for "To Go")=IFERROR(IF(ABS(L{n}/J{n})>5,"n.m.",L{n}/J{n}),"")=IFERROR(IF(ABS(V{n}/H{n}-1)>5,"n.m.",V{n}/H{n}-1),"")Read the blueprint's formatting sections carefully. Most critical rules:
Borders — full width:
Revenue row special treatment (Output Tab):
#E0DBD7) on data columns only (B–H, J–M, O–R, T) — NOT on spacer columnsNumber formats:
| Type | Format |
|---|---|
| Subsegment data | _(#,##0.0_)_%;(#,##0.0)_%;_("–"_)_%;_(@_)_% |
| Total dollar rows | _([$]#,##0_)_%;([$]#,##0)_%;_("–"_)_%;_(@_)_% |
| Percentage rows | _(#,##0.0%_);(#,##0.0%);_("–"_)_%;_(@_)_% |
| Output Tab dollars | _(#,##0.0_)_%;(#,##0.0)_%;_("–"_)_%;_(@_)_% |
| Output Tab percentages | _(#,##0.0%_);(#,##0.0%);_("–"_)_%;_(@_)_% |
After saving:
python scripts/recalc.py output.xlsx 120
Check for errors in the 3 new tabs only (ignore pre-existing source tab errors)
Spot-check values: Compare 2–3 FMT cells against the source P&L to confirm formulas pull correctly
Verify formatting: Confirm gridlines off, borders on totals, fills correct
Fix any errors and recalculate again before delivery
Save to outputs folder and provide a download link.
Column offset errors: When mapping FMT columns to source P&L columns, double-check the offset. The INDEX/MATCH with row 1 dates handles this — but verify with a spot check.
Missing Output Tab formulas: Every segment and subsegment needs SUMIFS data formulas — not just labels.
Gridlines off = borders critical: Any missing border creates a visible gap. Full-width borders on total rows and consistent vertical borders on the Output Tab are essential.
Division by zero: Always wrap percentage formulas in IF or IFERROR guards.
=-+ syntax: LibreOffice doesn't support =-+value. Use =-value instead.
Variance "n.m." handling: When variance percentage exceeds 500% (|x| > 5), display "n.m." instead of the number.
Adaptive categories: If the source has more than 5 expense categories, add them — don't force data into S&M / R&D / G&A if that's not what the source uses.
npx claudepluginhub ian-lawrence423/claude-skills --plugin deal-diligenceProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.