financial-advisor

Files (1)

• SKILL.md

SKILL.md (first lines)

Generate a monthly personal financial analysis using 4 parallel agents (spending analyzer, debt strategist, credit health monitor, budget/savings advisor).

Target: $ARGUMENTS

## Constants

```
FINANCIAL_ROOT = $ONEDRIVE_PERSONAL/Personal/Financial Reports
PERSONAL_ROOT  = $ONEDRIVE_PERSONAL/Personal
```

## Step 0: First-Run Setup and Directory Verification

Check that all required subdirectories exist under `FINANCIAL_ROOT`:

```
Bank Statements/
Credit Card Statements/
Credit Score/
Debt Tracking/
Mortgage/
Investments/
Pay Stubs/
Insurance/
Loan Forgiveness/
Other/
Reports/
```

If any are missing, create them with `mkdir -p`.

### Mode Routing

- If `$ARGUMENTS` is `--setup`: create directories, print confirmation, and STOP.
- If `$ARGUMENTS` is `--migrate`: run the migration script (Step 0b) and STOP.
- Otherwise: proceed to Step 1.

### Step 0b: File Migration (--migrate only)

Write and execute `/tmp/financial_migrate.py`. The script copies (NOT moves) files from scattered locations into the organized structure using `shutil.copy2` to preserve timestamps. Skip files that already exist at the destination (compare filename only).

**Print a dry-run summary first, then execute.**

Source-to-destination mapping:

| Source | Destination |
|--------|-------------|
| `PERSONAL_ROOT/Bank Statements/Paystub/*.pdf` | `Pay Stubs/` |
| `PERSONAL_ROOT/Banking/stmt.csv` | `Bank Statements/` |
| `PERSONAL_ROOT/Banking/transactions.xlsx` | `Bank Statements/` |
| `PERSONAL_ROOT/Banking/eStmt_*.pdf` | `Bank Statements/` |
| `PERSONAL_ROOT/Banking/Mortgage - Escrow Docs/*` | `Mortgage/` |
| `PERSONAL_ROOT/House/eStmt_*.pdf` | `Bank Statements/` |
| `PERSONAL_ROOT/CreditCardStatement.pdf` | `Credit Card Statements/` |
| `PERSONAL_ROOT/Debt.xlsx` | `Debt Tracking/` |
| `PERSONAL_ROOT/Debt_Amortization_Schedule.csv` | `Debt Tracking/` |
| `PERSONAL_ROOT/Out of Debt.xlsx` | `Debt Tracking/` |
| `PERSONAL_ROOT/House/Edward Jones*.pdf` | `Investments/` |
| `PERSONAL_ROOT/House/Edward Jones*.xlsx` | `Investments/` |
| `PERSONAL_ROOT/House/Liquidity of Roth IRA.pdf` | `Investments/` |
| `PERSONAL_ROOT/House/XXXX7338*edj*.pdf` | `Investments/` |
| `PERSONAL_ROOT/Stonks/Indices.xlsx` | `Investments/` |
| `PERSONAL_ROOT/Retirement/*.pdf` | `Investments/` |
| `PERSONAL_ROOT/Deerfield Bonds.pdf` | `Investments/` |
| `PERSONAL_ROOT/American Fidelity/*.pdf` | `Insurance/` |
| `PERSONAL_ROOT/Loan Forgiveness/*` | `Loan Forgiveness/` |
| `PERSONAL_ROOT/Credit Report/*.pdf` | `Credit Score/` |
| `PERSONAL_ROOT/House/*W2*.pdf` | `Pay Stubs/` |
| `PERSONAL_ROOT/House/*W2*.png` | `Pay Stubs/` |
| `PERSONAL_ROOT/House/*W2*.png.pdf` | `Pay Stubs/` |
| `PERSONAL_ROOT/House/*paystub*` | `Pay Stubs/` |
| `PERSONAL_ROOT/House/*Paystub*` | `Pay Stubs/` |
| `PERSONAL_ROOT/House/$AUTHOR_NAME CHECK STUBS.pdf` | `Pay Stubs/` |
| `PERSONAL_ROOT/House/employer-stub.pdf` | `Pay Stubs/` |
| `PERSONAL_ROOT/Child Support.xlsx` | `Other/` |
| `PERSONAL_ROOT/Personal - Travel.xlsx` | `Other/` |

After migration, print count of files copied per destination folder.

## Step 1: Determine Target Month

- If `$ARGUMENTS` matches `YYYY-MM` format, use that as target month.
- Otherwise, default to the current month (today's date).
- Set variables: `MONTH` (01-12), `YEAR` (YYYY), `MONTH_YEAR` (e.g., "February 2026"), `MONTH_TAG` (e.g., "2026-02").

## Step 2: Inventory Available Files

Scan each subdirectory for files relevant to the target month. Financial files may not always be month-tagged, so inventory everything available:

1. **Bank Statements/**: CSVs (`stmt.csv`, `transactions.xlsx`), PDF statements (`eStmt_*.pdf`)
2. **Credit Card Statements/**: PDF statements
3. **Debt Tracking/**: `Debt.xlsx`, `Out of Debt.xlsx`, `Debt_Amortization_Schedule.csv`
4. **Credit Score/**: Equifax.pdf, Experian.pdf, TransUnion.pdf
5. **Pay Stubs/**: Monthly paystub PDFs
6. **Investments/**: Edward Jones files, retirement statements
7. **Mortgage/**: Escrow documents

Print a summary table of available files. Warn if critical files are missing (bank statements, debt tracking).

Create the output directory: `FINANCIAL_ROOT/Reports/YYYY-MM/`

## Step 3: Create Team and Tasks

1. `TeamCreate` with `team_name: "financial-YYYY-MM"` (e.g., `financial-2026-02`).

2. `TaskCreate` for 4 tasks:

### Task 1: Spending Analysis
Reads: Bank Statements/ and Credit Card Statements/
- Parse CSV/XLSX files with pandas for transaction data
- Extract text from PDF statements with PyMuPDF (`fitz`) for amounts and descriptions
- Categorize every transaction using the keyword engine (see Categorization Engine below)
- Calculate: total income, total expenses, category totals, top 10 merchants
- Detect recurring charges (same merchant, similar amount, monthly pattern)
- Flag transactions over $500 as "large"
- Write results to `/tmp/financial_spending_YYYY-MM.json`

### Task 2: Debt Analysis
Reads: Debt Tracking/

From jasonbhorne/claude-code-skills, MIT.

More from jasonbhorne/claude-code-skills