Moody's Corporation (NYSE: MCO) has released its first library of AI skills, marking a shift from simply feeding data into AI agents to packaging the firm's analytical know-how into reusable, platform-agnostic instruction kits. The skills launch first on Microsoft 365 Copilot Cowork, with availability set to expand across other compatible AI platforms.
From Data Feeds to Decision-Grade Skills
Moody's describes its new offering as purpose-built, platform-agnostic instruction kits that encode its analytical frameworks and connect AI agents directly to its decision-grade intelligence. Rather than asking analysts to manually compile reports, customers can now execute complex analytical workflows through a single natural-language request, with outputs grounded in Moody's proprietary ratings, research, and risk data.
The first wave covers five high-priority financial workflows: Earnings Call Summary, which extracts revenue trends, pricing dynamics, consumer health, and tariff exposure from transcripts; Peer Analysis, which builds investor-grade comparisons across leverage, profitability, ESG, and credit quality; Public Information Book, a full dossier spanning financials, governance, and competitive positioning; Rating Pitch, a structured deck covering sector context and rating history; and Sector Analysis, which blends Moody's proprietary research with live market intelligence for a sector-level outlook.
"Moody's is among the first financial data providers to deliver a full library of skills on an open standard, and today's launch is just the beginning. AI platforms are becoming the interface for financial decision-making, and the next phase of adoption will be defined by execution."— Cristina Pieretti, Head of Digital Content and Innovation, Moody's
The Architecture Behind the Skills
A skill defines how the work gets done; Moody's Model Context Protocol (MCP) servers connect that skill to the underlying data it runs on. MCP is the open standard that lets an AI agent draw directly on Moody's ratings, research, and risk intelligence, ensuring outputs are grounded in proprietary data rather than general-purpose web content.
Each skill is captured in a simple, shareable instruction file built on the open SKILL.md format, which originated with Anthropic and has since been adopted by platforms including OpenAI, Microsoft, Google, and Amazon. Because the standard is open, the institutional knowledge encoded in each skill becomes a durable, portable asset rather than a capability locked to a single provider — built once, and able to run on any compatible platform.