Workday and Anthropic Connect Claude to Sana HR and Finance Agents

Workday and Anthropic on October 13, 2026, announced a partnership at Workday Rising 2026 in Las Vegas that connects Claude’s reasoning with Sana AI agents and Workday’s data and business context in HR and finance. The experience runs in either Sana, Workday’s agentic experience and orchestration layer, or in Claude, depending on the work being done.

How the Connection Works

Under the partnership, people can start work in Sana or in Claude with their context and controls intact in each. Claude and Sana exchange information and tasks through the Model Context Protocol (MCP), an open standard, so work carries forward rather than starting over.

In Sana, agents are orchestrated through complex HR and finance work, such as a compensation cycle or the financial close. Workday said Sana turns processes that previously took weeks into ones that happen continuously behind the scenes, holding the context of each process from start to finish and building the workspace each task needs, such as a live compensation grid or an exceptions queue, without moving data out of Workday. When a step calls for it, such as drafting the note a manager shares to explain an employee’s raise, Sana calls on Claude’s reasoning and shares only the context that step needs, while exceptions go to a person for review.

Inside Claude, employees can access Sana AI agents, analyze Workday HR and finance data, and take action. A hiring manager, for example, can compare open roles against existing headcount and budget, then build an internal mobility plan that fills those roles with people already on the team. A finance leader can check invoice exceptions against purchase orders, then approve them in Claude or route them to Sana. Every action follows the customer’s existing permissions and audit controls in Workday, and Claude surfaces only what that person is permitted to see.

Workday said joint customers will also be able to build their own agents directly in Claude using Workday Agent-Ready Tools, with the same governance, security, and audit trail built in from the start. Administrators connect once, with no new permission model to build and no per-user setup.

Gerrit Kazmaier, Workday’s president of product and technology, said customers want domain AI agents for their people and money that understand the business rules they run on. “Together, Workday Sana and Claude unlock lawful AI across the enterprise to get real work done,” he said.

A Relationship Already in Place

The partnership extends an existing mutual relationship. Anthropic is a Workday customer, running its finance and people operations on Workday Financials and Workday Human Capital Management, and relies on the same Workday governance, rules, approvals, and audit controls to help run its own business.

Workday also builds with Claude. The company has rolled out Claude Enterprise and plans to start using Cowork, and its product and development teams use Claude Code to build Workday’s products. Workday Developer Agent connects to Claude to help developers build on Workday. Workday said the Sana in Claude experience brings Sana and Workday Agent-Ready Tools to joint customers in Claude with its permissions and controls intact.

Ankur Sinha, head of engineering, enterprise and verticals at Anthropic, said in the announcement: “HR and finance decisions shape every employee’s pay, role, and career.” He said Claude brings the reasoning to help people make those decisions faster and with more confidence, with customers’ own controls applied at every step.

Availability and Platform Context

The Sana in Claude plugin is available to early access customers, and Workday expects it to be generally available in late 2026. Workday Agent-Ready Tools are generally available to all customers through Workday Extend AI and will be available through the plugin later in 2026. The companies said they will continue to collaborate on the roadmap for Sana and Claude, expanding the ways the two experiences work together as customer needs evolve.

The announcement came the same day Workday said its Trusted Agent Platform is headless, enabling customers to build, run, and govern AI agents across HR and finance with the development tools and models they choose. In that release, Workday described Agent-Ready Tools as its MCP tools, giving any agent that supports MCP an approved way to read Workday data and take action, and listed Agent Harness, Workday Data Cloud, Developer Agent, and Custom Agents as generally available. Teams building on the platform can choose which AI model powers each agent, with Anthropic’s Claude, Google’s Gemini, Meta’s Llama, Mistral models, OpenAI’s GPT, and NVIDIA’s Nemotron open models among the options. Sinha said in that release that MCP gives agents a standard way to connect with the systems where work happens and that Agent-Ready Tools preserve the permissions and approvals businesses already trust.

Workday said Agent Context Service, which will identify the Workday context an agent needs before it starts reasoning, is expected in early access within Agent Harness in late 2026. Grounded Reasoning, which will check requests against company policies, business processes, and tenant data before an agent acts, is expected within Agent Harness in early access in early calendar year 2027.

Workday Rising’s October 13 digital program includes the Office of the CIO keynote at 11:00 a.m. PT and a session titled The Anthropic Blueprint for Hypergrowth Finance at 1:45 p.m. PT. Workday states that more than 11,500 organizations worldwide, including more than 65% of the Fortune 500, trust it to deliver.