One day after announcing a $1.5 billion joint venture with Blackstone, Goldman Sachs, and Hellman & Friedman, Anthropic shipped what that venture will actually sell. On May 5, the company launched a library of roughly ten pre-built AI agent templates aimed squarely at the most labor-intensive work in financial services: pitchbooks, credit memos, KYC screening, month-end close, earnings analysis, underwriting, and insurance claims.

This is not a platform announcement. These agents are ready to run today.

What Anthropic Actually Released

The ten templates cover the work that financial analysts, compliance officers, and operations teams spend most of their time on. Each is designed to run as a plugin inside Claude Cowork or Claude Code alongside software analysts already use — no separate interface, no ripping out existing tools.

A few standouts:

The Pitch Agent connects to PitchBook, FactSet, and S&P Capital IQ to pull comparable companies, builds a comps model in Excel, drafts a pitchbook in PowerPoint, and writes a cover note in Outlook — all in one go, with context carried across every application simultaneously.

The KYC Screener assembles entity files from internal systems and external data sources, reviews source documents, and packages escalation cases for compliance review. What used to take an analyst a morning now takes minutes.

The Financial Crimes Agent, built in partnership with FIS, is aimed at anti-money-laundering investigations. It automatically assembles evidence across a bank’s core systems, evaluates activity against known typologies, and surfaces the highest-risk cases for human review. FIS says AML investigations that previously took hours can now be compressed to minutes.

The Microsoft 365 Integration Is the Real Story

Beyond the agents themselves, Anthropic announced full Microsoft 365 integration — Claude now works as a single agent across Excel, PowerPoint, Word, and Outlook simultaneously, carrying context between applications. Add-ins for all four became generally available today.

For anyone who works in finance, this matters. Analysts are not in one tool; they live across spreadsheets, slides, documents, and email in the same workflow. Claude now moves with them instead of requiring them to move into Claude.

The Data Partnerships

Anthropic connected Claude to a long list of financial data platforms: FactSet, S&P Capital IQ, MSCI, PitchBook, Morningstar, Chronograph, LSEG, and Daloopa. Moody’s took it a step further, embedding its full platform as a native Claude app so users can analyze credit ratings and risk data for more than 600 million companies without leaving the Claude interface.

That level of native integration is significant. The agent is not calling an API and pasting results — it is reasoning directly inside the data environment.

What This Means for Business

For data and finance professionals, this is the clearest signal yet that AI agents are moving from experiment to workflow infrastructure. The question for most organizations is no longer whether AI will touch financial analysis — it is whether they have the data foundations and AI literacy to put these tools to work effectively.

A few things to watch:

Vertical agents are replacing horizontal tools. General-purpose AI assistants are giving way to agents built specifically for a job. The firms winning the next wave are not building better chatbots; they are building agents that know the workflows, the data sources, and the edge cases of a specific domain.

Microsoft 365 is becoming the default AI interface for enterprise. Three of the four major productivity tools now have native Claude integration. For any business running on Office, AI is not coming — it is already in the dock.

Compliance and risk teams are the new early adopters. KYC, AML, and credit risk workflows have clear inputs, clear outputs, and measurable accuracy requirements. They are exactly the right environments for agentic automation to prove its value — and for organizations to build the governance muscle they will need when agents start touching more sensitive work.

For businesses not in financial services, the lesson still applies. Industry-specific, workflow-specific AI agents built on top of real data integrations are where the ROI is. The question is who in your organization is building them — and whether they have the data literacy to do it well.

---

Enterprise DNA helps data teams and business leaders build the skills and systems to work with AI at this level. Whether you are evaluating platforms, upskilling your analysts, or deploying agents in your own workflows, the path starts with strong data foundations.