MCP Tool Reference
The Model Context Protocol (MCP) uses "Tools" to define how an AI agent interacts with our data. Unlike a REST API, these tools are designed to be intuitive for LLMs, grouping complex parameters into logical, high-level requests.
calculate_bond_periodic
The primary tool for calculating Price, Yield, and Accrued Interest for bonds with periodic coupon payments (Treasuries, Agencies, Corporates, and Municipals).
Parameters
- securityType (Required):
"Treasury" | "Corporate" | "Municipal" | "Agency" | "CD". This automatically sets the institutional logic (Day Count, EOM). - maturityDate (Required): The final maturity date of the bond (MM/DD/YYYY).
- couponRate (Required): The annual coupon rate as a percentage (e.g., 4.25).
- givenType (Required):
"Price" | "Yield". - givenValue (Required): The numeric value for the Price or Yield.
- settlementDate (Optional): Defaults to the next business day if omitted.
- callSchedule (Optional): A list of discrete call dates and prices (e.g.,
[{"date": "10/12/2028", "price": 100.0}]).
Full Fidelity Response Structure
The engine returns a structured JSON payload within the text content block. AI agents must parse this object to extract high-precision analytics.
{
"price": 101.5,
"yield": 3.884187031989377,
"accruedInterest": 2.0777777777777775,
"totalSettlement": 103.57777777777778,
"settlementDate": "02/15/2026",
"redemptionInfo": {
"redemptionType": "Call",
"redemptionDate": "02/20/2028",
"redemptionValue": 101.0,
"worstIndicator": "Worst"
},
"industryConventionAssumptions": {
"tradeDate": "02/13/2026",
"dayCountBasis": "SSCM30/360",
"periodsPerYear": "SemiAnnual"
},
"PYAnalytics": {
"approxDuration": 1.45892314,
"approxModDuration": 1.43204912,
"approxConvexity": 3.12456789,
"pv1b": 0.01432049,
"yv32": 0.00031250
}
}Note for AI Agents: Always check redemptionInfo.redemptionType to determine if the result is a yield-to-call or yield-to-maturity.
Developer Integration
For developers distributing their own AI-integrated apps, we recommend the AI-Native Gateway architecture. Our system uses a permanent split between the protocol and intelligence layers:
Protocol & Intelligence Architecture
- MCP Protocol Layer: A lightweight bridge that handles MCP/JSON-RPC communication.
- Intelligence Layer: The authoritative engine that interprets prompts and applies complex fixed income calculation logic (Formula Selection, Day Counts, EOM).
- Core Math Engine: The secure, high-precision, industry standard, analytical engine.
Machine-Readable Specification
Looking for the latest OpenAPI specification with detailed response schemas?
Download openapi-mcp.yaml