Anthropic Launches Ready-to-Run Claude Agent Templates for Financial Services

May 5, 2026 · Updated May 13, 2026

Anthropic launched ready-to-run agent templates for the financial services industry, providing pre-configured "digital employees" for tasks like building pitch books and closing month-end books. These templates package specific agent skills (reusable, domain-specific capabilities), connectors, and subagents required to execute regulated workflows from start to finish.

Deployment options: Plugins, Managed Agent cookbooks
Microsoft 365 support: Excel, PowerPoint, Word, and others
Data connectors: FactSet, LSEG, Moody’s, and others
Compliance standards: SOC 2, FedRAMP
Primary use cases: Investment banking, insurance, asset management, and others

This move shifts Claude from a general assistant to a specialized vertical tool, addressing data grounding (connecting AI to verifiable information) and compliance needs of large institutions. The reliability of these agents mirrors Anthropic's Model Spec Midtraining, which reduces misalignment while building on the Claude Managed Agents infrastructure.

You can install these templates as plugins that extend Claude Cowork or Claude Code, or use provided cookbooks to run them as Managed Agents. They feature native integrations with Excel and PowerPoint, alongside connectors for data from FactSet, LSEG, Moody’s, and others.

View the full update on anthropic.com

Claude

@claudeaiMay 5

New for financial services: ready-to-run Claude agent templates for building pitches, conducting valuation reviews, closing the books at month-end, and more. Install them as plugins in Cowork and Claude Code, or use our cookbooks to run them in production as Managed Agents. https://t.co/Nzj4Feaaz2

2.9k34k

View on X

Still wondering? A few quick answers below.

These are pre-packaged AI agent configurations designed to automate specific workflows in banking, insurance, and asset management. Unlike general chat prompts, these templates include the necessary skills, data connectors, and subagents to perform complex tasks like building valuation models, conducting credit underwriting, and performing KYC screenings from start to finish.

You can deploy these templates in three primary ways depending on your technical needs. They are available as plugins within the Claude Cowork and Claude Code applications for immediate use. For production-grade integration, developers can use Anthropic's cookbooks to run them as Managed Agents on the Claude platform infrastructure.

Anthropic has built native connectors that allow Claude to reason over data from industry-standard providers. These include LSEG, FactSet, Moody’s, and others. These integrations ensure that the agents operate on verifiable, professional-grade market data and research rather than relying solely on the model's internal training data.

Yes, these financial agents are designed to work natively within Microsoft 365 applications. Claude can read formulas and process data directly in Excel, generate and edit slides in PowerPoint, and draft memos in Word. This allows investment professionals to build pitch books and financial models within the tools they already use daily.

Anthropic has built these solutions to meet the rigorous requirements of financial risk committees. The infrastructure supports SOC 2 and FedRAMP compliance. Additionally, every output generated by the agents is source-attributed, allowing human teams to verify the data and logic behind every number before taking action or finalizing a report.