Cursor Task: Normalize Snapshots to $500/Month

Context

The SD App snapshot JSONs currently use test fixture loan amounts (interest-only: $100,000; term loan: $37,543). These were chosen to validate math correctness, not for user-facing presentation.

The user-facing standard has been set: $500/month of gross (before-tax) investable cash flow. All snapshot outputs should reflect this normalization.

Interest-only: loan_amount such that annual gross interest = $6,000 → $66,667 at 9%
Term loan: loan_amount such that annual payment = $6,000 → $38,506 at 9% for 10 years; loan_payment = $6,000

This is a pure Python task — update two snapshot generator files, regenerate two JSON files. No API changes, no SvelteKit changes.

Files to Read First (in order)

packages/sd-math/tools/generate_snapshot.py — interest-only generator
packages/sd-math/tools/generate_snapshot_term_loan.py — term loan generator
apps/sd-app/static/snapshots/int-only-ca.json — current output (for reference)
apps/sd-app/static/snapshots/term-loan-ca.json — current output (for reference)
packages/sd-math/src/sd_math/models/inputs.py — LoanProfile, TermLoanProfile definitions
packages/sd-math/src/sd_math/strategies/interest_only.py — to understand at_annual_investment derivation
packages/sd-math/src/sd_math/strategies/term_loan.py — to understand loan_payment handling

Hard Rules

Do NOT change sd-math library code — only the generator scripts and the output JSON files
Do NOT change the API (packages/sd-api/)
Do NOT change SvelteKit files (apps/sd-app/)
Do NOT change test fixtures (packages/sd-math/tests/fixtures/) — those are for math validation, not UX normalization
Field names in the output JSONs must remain identical to current — only the values change
Run the generators using the project’s Python environment (same as sd-math tests): cd packages/sd-math && python tools/generate_snapshot.py etc.

Changes Required

1. `packages/sd-math/tools/generate_snapshot.py`

Change the LOAN definition:

# BEFORE
LOAN = LoanProfile(
    loan_amount=100_000.0,
    ...
)

# AFTER
LOAN = LoanProfile(
    loan_amount=66_667.0,   # $500/month gross: $66,667 × 9% = $6,000/year
    ...
)

Everything else in the file stays the same. The generator will produce new output values automatically.

2. `packages/sd-math/tools/generate_snapshot_term_loan.py`

Change the TERM_LOAN definition:

# BEFORE
TERM_LOAN = TermLoanProfile(
    loan_amount=37_543.0,
    loan_payment=5_850.0,
    loan_interest_rate=0.09,
    n_loan_periods=10,
)

# AFTER
TERM_LOAN = TermLoanProfile(
    loan_amount=38_506.0,   # $500/month gross: annual payment = $6,000
    loan_payment=6_000.0,   # $500/month × 12 = $6,000/year
    loan_interest_rate=0.09,
    n_loan_periods=10,
)

Everything else in the file stays the same.

Deliverables

Both generator files updated with new loan amounts
Both JSON snapshots regenerated:
- apps/sd-app/static/snapshots/int-only-ca.json
- apps/sd-app/static/snapshots/term-loan-ca.json

Verification

After regenerating, confirm these values in the new JSON files:

int-only-ca.json — middle-income profile:

inputs.loan_amount = 66667
analysis_summary.at_annual_investment ≈ 3900 (within ±50) (66,667 × 9% = $6,000 gross, minus 35% tax benefit = $3,900 net)
analysis_summary.bt_annual_interest = 6000 (exactly)

term-loan-ca.json — middle-income profile:

inputs.loan_amount = 38506
inputs.loan_payment = 6000
analysis_details.annual_rows[0].interest_part_of_pmt ≈ 3465 (Year 1: 38,506 × 9%)
analysis_details.annual_rows[0].loan_reduction ≈ 2535 (Year 1: 6,000 - 3,465)

Completion Signal

Paste the output of both generator runs here:

# Run 1 — Interest-only
cd packages/sd-math && python tools/generate_snapshot.py
# [paste output]

# Run 2 — Term loan
cd packages/sd-math && python tools/generate_snapshot_term_loan.py
# [paste output]

Then paste the analysis_summary block from each new JSON file (middle-income profile only) so the values can be spot-checked.

Completion Record — 2026-03-22

Generator run output

# Run 1 — Interest-only
cd packages/sd-math && python tools/generate_snapshot.py

✓ Snapshot written to /home/ta/projects/monorepo/apps/sd-app/static/snapshots/int-only-ca.json
  Profiles: ['middle-income', 'high-income']
  middle-income: 5 summary scenarios, 10 annual rows
  high-income: 5 summary scenarios, 10 annual rows

# Run 2 — Term loan
cd packages/sd-math && python tools/generate_snapshot_term_loan.py

✓ Snapshot written to /home/ta/projects/monorepo/apps/sd-app/static/snapshots/term-loan-ca.json
  Profiles: ['middle-income', 'high-income']
  middle-income: 5 summary scenarios, 10 annual rows
  high-income: 5 summary scenarios, 10 annual rows

Verification — int-only-ca.json (middle-income)

Check	Expected	Actual	Pass?
`inputs.loan_amount`	66667	66667	✓
`analysis_summary.at_annual_investment`	≈ 3900 (±50)	3900	✓
`analysis_summary.bt_annual_interest`	6000	6000	✓

analysis_summary block:

{
  "at_annual_investment": 3900,
  "bt_annual_interest": 6000,
  "scenarios": [
    { "expected_return": 0.0,    "no_lev_bt_balance": 39000, "net_lev_bt_bal": 0,     "net_increase": -39000, "pct_increase": -1.0,    "is_better_than": false },
    { "expected_return": 0.03,   "no_lev_bt_balance": 44391, "net_lev_bt_bal": 19413, "net_increase": -24979, "pct_increase": -0.5627, "is_better_than": false },
    { "expected_return": 0.0666, "no_lev_bt_balance": 52184, "net_lev_bt_bal": 52184, "net_increase": 0,      "pct_increase": 0.0,     "is_better_than": true  },
    { "expected_return": 0.07,   "no_lev_bt_balance": 52972, "net_lev_bt_bal": 55759, "net_increase": 2787,   "pct_increase": 0.0526,  "is_better_than": false },
    { "expected_return": 0.1,    "no_lev_bt_balance": 60636, "net_lev_bt_bal": 92733, "net_increase": 32097,  "pct_increase": 0.5293,  "is_better_than": false }
  ]
}

Verification — term-loan-ca.json (middle-income)

Check	Expected	Actual	Pass?
`inputs.loan_amount`	38506	38506	✓
`inputs.loan_payment`	6000	6000	✓
`annual_rows[0].interest_part_of_pmt`	≈ 3465 (38,506 × 9%)	3466	✓
`annual_rows[0].loan_reduction`	≈ 2535 (6,000 − 3,465)	2534	✓