# Example Output ## Scenario User asks: "Model a 1,000-claim portfolio with a CFA and funder MOIC. I have rough assumptions but no spreadsheet." Access state: specification/calculation mode. Claude can do arithmetic but has no validated damages distribution or legal enforceability advice. ## Good Output Shape ```yaml access_mode: calculation_mode valuation_date: "2026-05-10" currency: GBP tax_vat_basis: "excluded unless stated" real_or_nominal: nominal assumptions: - name: claim_count value: 1000 source: user_supplied - name: damages_mean value: 12000 currency: GBP source: user_supplied - name: win_probability value: 0.58 source: assumption_to_verify - name: funder_moic value: 2.0 source: user_supplied - name: funded_costs value: 1800000 currency: GBP source: user_supplied calculated_outputs: - metric: expected_gross_recovery value: 6960000 currency: GBP formula: "claim_count * damages_mean * win_probability" calculation: "1000 * 12000 * 0.58" confidence: medium - metric: funder_entitlement_before_caps value: 3600000 currency: GBP formula: "funded_costs * funder_moic" calculation: "1800000 * 2.0" - metric: residual_before_firm_fees_and_client_distribution value: 3360000 currency: GBP formula: "expected_gross_recovery - funder_entitlement_before_caps" caveat: "Does not include solicitor fees, ATE, adverse costs, tax/VAT, or distribution waterfall beyond funder entitlement." confidence: low not_modelled: - "Funding enforceability not assessed." - "CFA/DBA regulatory compliance not assessed." - "No Monte Carlo percentiles produced because no damages/time distribution and iteration method supplied." next_inputs_needed: - "Costs budget by phase." - "ATE premium and adverse-cost assumptions." - "Waterfall priority and caps." - "Damages distribution if stochastic modelling is required." ``` ## Bad Output To Avoid ```text This portfolio is commercially viable. ``` Model economics; do not recommend bringing, funding, settling, or abandoning a claim.