AI-powered financial intelligence and agent service with MCP tools for financial calculations and document-based Q&A
Three tools with complete input schemas and clear descriptions. All parameters are typed and described. However, descriptions are formulaic and lack LLM-optimized guidance on when/why to use each tool. No output schemas documented. No error handling guidance. Tool names follow verb_noun pattern correctly but lack distinction for domain context. Missing idempotency hints and composition guidance.
Calculates the future value of a lump sum investment using compound interest, given principal, annual interest rate, time in years, and compounding frequency per year.
Calculates the monthly EMI (equated monthly installment) for a loan, given the principal amount, annual interest rate, and tenure in years.
Calculates the future value of a monthly SIP (systematic investment plan), given the monthly investment amount, expected annual return rate, and duration in years.
No output schemas documented. LLMs cannot plan downstream operations or extract return values. Each tool should document: return type (double/object), field names, and units (e.g., 'Returns: {emi: double (monthly payment in currency units), totalPayment: double, totalInterest: double}').
Descriptions lack LLM-optimized guidance. Current: 'Calculates the monthly EMI...' (generic). Better: 'Use this to estimate monthly loan payments. Requires principal (loan amount), annual interest rate (e.g., 8.5 for 8.5%), and tenure in years. Returns monthly payment amount. Useful for comparing loan offers or budgeting.'
No error handling guidance. What happens if principal is negative? If tenure is 0? If compoundingFrequency exceeds 365? Tools should validate inputs and return actionable errors: 'Invalid tenure: must be 1-50 years' not a 500 error.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | 2026-07-28+ | v2 |
Parameter descriptions lack format/range constraints. 'Annual interest rate as a percentage, e.g. 8.5 for 8.5%' is good but incomplete. Add: 'Range: 0.1 - 30%. If you have a decimal rate (0.085), multiply by 100 first.'
No composition guidance. If an agent wants to compare loan offers, it must call calculateEmi three times manually. Consider a batch tool (calculateEmiComparison) or document the multi-call pattern in descriptions.