Breaking

Anthropic Releases 10 Claude Agents for Financial Workflows on GitHub

June 29, 2026 at 15:19 EDT

AI Agents
Finance
Sales & Business

Anthropic has published 10 ready-to-run Claude agent templates that automate pitch deck creation and market research for investment banking and research teams. Announced on May 5, 2026 as "Agents for financial services," the templates are available in a public GitHub repository. According to the official announcement, they run as plugins for Claude Cowork and Claude Code, or as a cookbook for Claude Managed Agents.

May 5, 2026 · Anthropic

10 Claude Agents Built to Run Wall Street Workflows

Anthropic ships ready-to-run agent templates that automate time-consuming finance tasks — from pitchbook creation and KYC screening to month-end close — packaged as plugins, cookbooks, and cloud-hosted Managed Agents.

ready-to-run agent templates

building blocks: skills · connectors · subagents

64.37%

Claude Opus 4.7 on Vals AI Finance Agent benchmark

The 10 agents, split across two domains

Research & Client Coverage

Pitch builder — target lists, comparables, pitchbooks
Meeting preparer — meeting prep
Earnings reviewer — earnings review
Model builder — financial model building
Market researcher — market research
Valuation reviewer — valuation checks

Finance & Operations

General ledger reconciler — ledger reconciliation
Month-end closer — month-end close
Statement auditor — financial statement audit
KYC screener — KYC screening

How one agent is assembled

Each template is a reference architecture combining three elements.

Skills

instructions + domain knowledge

Connectors

governed access to data

Subagents

extra Claude models for subtasks

→

Agent template

runs in production

Connectors span FactSet, S&P Capital IQ, PitchBook, Morningstar, and Moody's — with Microsoft 365 add-ins (Excel · PowerPoint · Word) preserving context from model to deck.

Early traction

Walleye Capital adopted Claude Code firm-wide; Carlyle and FIS cite shorter AML investigations and productivity gains. Developers report "operating like a team of specialists."

Still to verify

Governance and audit logs are critical in finance. Detailed reviews of actual deliverables remain limited — real-world effectiveness awaits further verification.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.