--- title: "Anthropic's 10 Pre-Built Finance Agent Templates: Pitchbooks, KYC, Close" description: "Anthropic unveiled 10 pre-built finance agent templates on May 5, 2026 across pitchbook building, KYC screening, and month-end close. What each template does and the hours it replaces." canonical: https://callsphere.ai/blog/tw26w19-anthropic-10-finance-agent-templates-pitchbook-kyc-month-end category: "AI Engineering" tags: ["Anthropic", "Claude Cowork", "Claude Code", "Finance Agents", "Pitchbook AI", "KYC Automation", "Month-End Close"] author: "CallSphere Team" published: 2026-05-06T00:00:00.000Z updated: 2026-05-11T04:30:37.631Z --- # Anthropic's 10 Pre-Built Finance Agent Templates: Pitchbooks, KYC, Close > Anthropic unveiled 10 pre-built finance agent templates on May 5, 2026 across pitchbook building, KYC screening, and month-end close. What each template does and the hours it replaces. ## The Announcement At a New York City briefing on May 5, 2026, Anthropic introduced **ten pre-built finance agent templates** for Claude Cowork and Claude Code, with matching cookbooks for Claude Managed Agents. The templates target pitchbook building, KYC screening, and month-end close. They are available to banks, asset managers, and insurers. The templates land alongside a Moody's data partnership, full Microsoft 365 integration, and a Vals AI benchmark result of 64.37 percent for Claude Opus 4.7 on finance agent work. This piece walks through the ten templates, the analyst hours each one replaces, and what changes for a typical finance team. ## Why Templates, Not Just A Model A frontier model alone does not deliver an end-to-end finance workflow. It needs prompts, tools, document layouts, validation steps, and stopping conditions. That work is repeated by every bank that tries to adopt LLMs. Anthropic packaging ten of the most common workflows as templates is the same move that turned spreadsheets into Excel templates twenty-five years ago. Templates compress weeks of internal engineering into hours. ## The Ten Templates The ten templates cluster into three families. ### Pitchbook And Deal Materials 1. **Pitchbook builder.** Ingests company filings, comparable company data, and analyst commentary. Produces a first-draft pitchbook with footnoted exhibits. Replaces a 20 to 40 hour associate task with a 1 to 2 hour review. 2. **Comparable companies update.** Refreshes a comp set with current trading multiples, segment splits, and recent transactions. Replaces 3 to 5 hours of associate work per refresh. 3. **Management presentation prep.** Drafts board and management decks from a structured outline plus source filings. Replaces 5 to 10 hours per deck. ### KYC, Compliance, And Screening 1. **KYC narrative drafting.** Compiles ownership structure, sanctions screening, and adverse media into a compliance-ready narrative. Replaces 2 to 4 hours per file. 2. **Enhanced due diligence.** Extends KYC with deeper source review and structured risk scoring. Replaces 6 to 12 hours per file. 3. **Transaction monitoring triage.** Reviews flagged transactions, drafts disposition notes, and routes for analyst confirmation. Replaces around 20 to 30 minutes per alert at scale. ### Month-End Close And Reporting 1. **Account reconciliation.** Matches subledger to general ledger, flags variances above threshold, drafts narratives. Replaces several days of accountant work per close. 2. **Flux analysis.** Explains period-over-period movements in P&L and balance sheet lines with source citations. Replaces 4 to 8 hours per close. 3. **Disclosure drafting.** Drafts MD&A sections and footnote candidates from underlying data plus prior period filings. Replaces 8 to 20 hours per quarter. 4. **Audit support package.** Assembles audit-ready support documents and ties figures back to source systems. Replaces a recurring 1 to 3 day audit prep task. The hours-replaced numbers above are typical bank ranges from public Anthropic and customer commentary. Actual savings depend on existing tooling and review policy. ## What The Customer List Tells You Anthropic has been clear that **JPMorgan Chase, Goldman Sachs, Citi, AIG, and Visa** are using Claude in production, with the broader financial services push starting in July 2025. Templates are the natural next layer for that customer base. A name-brand bank does not buy a model; it buys a workflow that already maps to its desk. The presence of insurers (AIG) and a payment network (Visa) in the customer list is what makes the template list broad rather than only investment banking. KYC, transaction monitoring, and close are universal across these institutions. ## Where Claude Cowork And Claude Code Fit Claude Cowork is the collaborative workspace for knowledge workers. Claude Code is the engineering surface. The same template can run in either: - An associate uses **Cowork** to drive the pitchbook template through a UI, with a human approving each major step. - A platform engineer uses **Claude Code** to wire the same template into an internal pipeline, version-controlled, and triggered by an event such as a new deal in the pipeline. The cookbooks for **Claude Managed Agents** extend the templates to long-running workflows that run on Anthropic infrastructure, suitable for overnight reconciliation runs or scheduled disclosure drafts. ## Where CallSphere Fits CallSphere is an AI voice and chat agent platform for customer-facing communication. We are not a finance back-office tool, but every bank, insurer, and asset manager with a customer-facing arm needs a 24/7 voice and chat front door. Where the new finance templates change CallSphere's conversation with financial services prospects: - The back office is finally getting reliable AI. Customers will start to expect the same from the front office. A voice agent that takes 30 minutes to call back is no longer competitive against a front-door AI that answers in two seconds. - HIPAA-friendly architecture and audited tool use map cleanly to financial services compliance reviews. - 57 plus languages mean a global insurer can launch one voice front-door instead of one per market. Pricing is Starter $149 per month for 2,000 interactions, Growth $499 for 10,000, Scale $1,499 for 50,000, with a free trial and 3 to 5 business day launches. See [pricing](https://callsphere.ai/pricing). ## What Changes For Finance Teams This Quarter Three concrete shifts to expect: 1. **First-draft work moves to the agent.** Associates and analysts spend more time editing and challenging the agent output, less time producing it. 2. **The bottleneck moves to data plumbing.** With Moody's data and Microsoft 365 integration available, the friction is now connecting internal systems, not the model. 3. **Review policies are the new differentiator.** Banks that define clear human-in-the-loop checkpoints will move faster than those that try to either fully automate or fully manually review. ## FAQ **Q: Are these templates open source?** The cookbooks for Claude Managed Agents are reference patterns. The Cowork and Code templates ship inside the Anthropic platform. **Q: Can a mid-market bank use these without a platform team?** Yes for the Cowork-driven templates. The Managed Agents and Code paths assume some engineering capacity. **Q: Does CallSphere integrate with these templates?** CallSphere lives at the customer-facing layer. Where a customer interaction surfaces a need for a back-office workflow (statement copy, dispute filing, claim status), CallSphere hands off through standard webhooks and CRM integrations. --- Source: https://callsphere.ai/blog/tw26w19-anthropic-10-finance-agent-templates-pitchbook-kyc-month-end