Workday and Anthropic announced a partnership on 13 October 2026 that links Workday’s Sana AI agents with Anthropic’s Claude, so people can start HR and finance work in either one and carry it across to the other [1]. Workday announced the deal at its Workday Rising event in Las Vegas. Neither company disclosed any financial terms.
MCP, in plain words
The two systems talk to each other through the Model Context Protocol, or MCP. The project’s own site describes MCP as an open-source standard for connecting AI applications to outside systems such as databases, tools and workflows, and compares it to a USB-C port for AI: one standard plug instead of a custom cable for every device [2]. Workday says that because Claude and Sana swap information and tasks over MCP, work “carries forward rather than starting over” when it moves between them [1].
How it works in each direction
Starting in Sana. Sana is Workday’s agent layer that runs long HR and finance jobs, such as a pay review cycle or closing the books at month end. According to Workday, when one step needs language or reasoning, for example drafting the note a manager uses to explain a pay rise, Sana calls on Claude and shares only the context that step needs, while the data stays inside Workday. Exceptions go to a person to review [1].
Starting in Claude. Employees will be able to use Sana agents, look at Workday HR and finance data and take action without leaving Claude. Workday’s examples include a hiring manager comparing open roles with headcount and budget, and a finance lead checking invoice exceptions against purchase orders and approving them [1].
Building your own. Joint customers can also build their own agents in Claude using Workday’s Agent-Ready Tools [1].
The permissions point is Workday’s claim
Workday says every action follows the customer’s existing permissions and audit controls in Workday, that Claude shows only what each person is allowed to see, and that the user confirms every change [1]. That is how Workday describes the design; the release does not point to any independent test of it. Workday also says admins connect once, with no new permission model and no per-user setup.
When it arrives
According to the release, the Sana in Claude plugin is in early access now, with general availability expected in late 2026. Workday’s Agent-Ready Tools are generally available now to all customers through Workday Extend AI, and Workday says they will reach the plugin later this year [1].
What the two companies said
Gerrit Kazmaier, Workday’s president of product and technology, said customers want domain AI agents “that understand the business rules they run on and leverage the extraordinary reasoning in Claude”, adding that together the two “unlock lawful AI across the enterprise to get real work done” [1].
Ankur Sinha, Anthropic’s head of engineering for enterprise and verticals, said: “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” [1].
Each is the other’s customer
The release says Anthropic runs its own finance and people operations on Workday Financials and Workday Human Capital Management (HCM). Going the other way, Workday has rolled out Claude Enterprise, and its product and development teams use Claude Code to build Workday’s products [1].
What this means
- One plug, two directions. By using MCP, an open standard, Workday is letting work flow both into Claude and out of it, rather than building a closed link.
- Trust is the pitch. The key selling point, that Workday’s rules and audit trail apply to every request, comes from Workday and has not been independently tested.
- Most of it is still coming. Only the Agent-Ready Tools are generally available today; the main plugin is in early access until late 2026.
- No money disclosed. The companies gave no financial terms.
Sources
- Workday Inc. via PR Newswire, “Workday and Anthropic Announce New Partnership to Connect Reasoning to the Rules That Run the Business, Wherever People Choose to Work”, 13 October 2026. https://www.prnewswire.com/news-releases/workday-and-anthropic-announce-new-partnership-to-connect-reasoning-to-the-rules-that-run-the-business-wherever-people-choose-to-work-302906752.html
- Model Context Protocol, “What is the Model Context Protocol (MCP)?”. https://modelcontextprotocol.io/docs/getting-started/intro
Related stories
- American Express Publishes “Intent Intelligence” to Stop AI Agents Buying What You Did Not Approve
- IBM Research Says Open AI Agents Sped Up a Quantum Decoder Chip Design 52 Times for About $26
- Workday to Acquire C1.ai, an Identity Platform for Managing What People and AI Agents Can Access
- Rocket Software Buys Geniez AI, a Six-Person Israeli Startup Building AI Agents for Mainframes
