Companies will soon be able to use Hermes Agent, the open-source artificial intelligence agent developed by Nous Research, directly inside Microsoft Teams. The integration is part of a new enterprise offering called Hermes Enterprise, announced on November 18, 2026. It allows employees to delegate tasks such as financial analysis, document drafting, research and inbox management through the workplace messaging platform.

The partnership brings together Nous Research’s AI agent technology and Microsoft’s enterprise software and cloud infrastructure. Beyond answering questions, Hermes is designed to remember how tasks are completed, turn successful workflows into reusable skills and share approved knowledge across an organization. The goal is to help businesses build AI assistants that become more useful as employees work with them.

Nous Research Partners With Microsoft to Launch Hermes Enterprise

Nous Research announced its partnership with Microsoft to bring Hermes Agent into the company’s enterprise ecosystem.

The new offering, Hermes Enterprise, is available through several Microsoft services, including Microsoft Teams, the Microsoft Store and Microsoft Azure Container Apps Sandboxes.

The integration allows employees to communicate with their AI agent through Teams rather than switching between separate applications. Workers can assign tasks through chat and receive the completed work in the same conversation.

Examples of supported tasks include:

  • Financial analysis and research.
  • Drafting business documents.
  • Organizing and triaging email inboxes.
  • Reading, analyzing and converting files.
  • Automating repetitive knowledge-work tasks.

The announcement also highlights Nous Research’s relationship with M12, Microsoft’s venture capital fund, which has invested in the AI company.

What Is Hermes Agent?

Hermes Agent is an open-source AI agent developed by Nous Research. Unlike a conventional chatbot that primarily responds to individual prompts, an AI agent can use tools, perform multiple steps and work toward completing a task.

Hermes is designed to retain useful working knowledge from previous tasks. When it discovers an effective way to complete a recurring activity, it can turn that process into a reusable skill.

For example, an employee might repeatedly ask Hermes to analyze financial reports using a particular structure. After learning the workflow, the agent can save the procedure and apply it to similar requests in the future.

This approach reduces the need for employees to explain the same instructions repeatedly.

Nous Research describes Hermes as a self-improving agent because its reusable skills and working memory can make recurring tasks more consistent over time. This does not mean the underlying AI model automatically retrains itself after every conversation. The improvement comes from retaining and reusing procedures and knowledge.

Collective Wisdom: How Hermes Shares Knowledge Across Teams

One of the most important features of Hermes Enterprise is called Collective Wisdom.

In many organizations, employees develop their own methods for completing tasks. However, these processes often remain limited to individual workers or departments.

For example, one employee may develop an efficient method for analyzing monthly sales reports, while another may create a useful process for preparing market research documents. Without a shared system, other employees may have to develop those workflows independently.

Collective Wisdom is designed to address this problem.

When Hermes identifies a useful skill, an employee can propose it for sharing with colleagues. After review and approval, the skill can become available to other team members.

This creates a shared collection of reusable workflows that employees can build on.

FeatureHow It Helps Businesses
Working memoryRetains useful context from previous tasks
Reusable skillsAutomates recurring activities
Collective WisdomShares approved workflows across teams
Microsoft Teams integrationAllows employees to delegate work through chat
Azure sandbox executionRuns agent workloads in isolated environments

Nous Research says the accumulated organizational knowledge remains a private company asset rather than being used as training data by an AI provider.

Hermes Agent Runs on Microsoft Azure

The partnership also includes integration with Microsoft Azure Container Apps Sandboxes.

AI agents can perform complex tasks involving files, tools and external services. This creates security challenges because an agent may need access to computing resources and information to complete its work.

Azure Container Apps Sandboxes provide isolated microvirtual-machine environments for executing agent workloads.

According to Nous Research, these environments can start in less than a second and release computing resources when they are idle.

Businesses can also run Hermes Enterprise within their own Azure tenant. This is particularly relevant for organizations with specific data residency, security and procurement requirements.

The approach gives companies another option for deploying AI agents while maintaining greater control over their cloud environments.

However, using an isolated execution environment does not automatically eliminate every security or privacy risk. Companies still need appropriate access controls, data policies and oversight of agent activities.

How Employees Can Use Hermes Inside Microsoft Teams

The integration is designed to fit into existing workplace communication processes.

An employee can open a conversation with Hermes and assign a task in natural language. The agent can then work on the request and return its results through Teams.

For example, an employee could ask Hermes to analyze a financial document, prepare a summary or draft a business report. Employees can also share files with the agent for analysis and conversion.

Nous Research’s Teams help documentation explains that users need a Nous account and an existing Hermes agent. The organization’s Teams administrator must also allow the application.

The setup involves adding the Hermes application to Teams, connecting the employee’s Teams account to their Nous account and linking the selected agent.

At launch, the Microsoft Marketplace listing describes support for private one-to-one conversations, while Nous Research’s subsequently updated help documentation also explains how to use the agent in group chats and channels through mentions. Availability may depend on the app version and organization configuration.

What the Partnership Means for Microsoft and the AI Industry

The partnership reflects the growing competition to bring AI agents into the software employees already use.

Microsoft has been expanding its Copilot ecosystem across applications such as Teams, Word, Excel and Outlook. However, Hermes Enterprise represents a different approach: it brings an independently developed open-source agent into Microsoft’s enterprise environment.

For businesses, this could create more flexibility in choosing AI agents for specific workflows instead of depending entirely on a single general-purpose assistant.

It also highlights a broader shift from conversational AI toward task-oriented systems. Businesses increasingly want AI tools that can research information, process documents, automate repetitive work and remember established procedures.

For Nous Research, the Microsoft partnership offers a route to reach enterprise customers through an established workplace software ecosystem.

For Microsoft, supporting external AI agents strengthens the appeal of Teams and Azure as platforms where businesses can deploy and manage AI-powered workflows.

The Bigger Picture

The most significant part of Hermes Enterprise may not be its ability to answer questions inside Microsoft Teams. It is the possibility of turning individual employees’ successful AI workflows into shared organizational knowledge.

If the system works reliably, businesses could reduce duplicated work and help employees reuse processes developed by colleagues. That could make AI agents more valuable over time, particularly in departments that repeatedly handle financial reports, research, documentation and administrative tasks.

However, enterprise adoption will depend on more than convenience. Reliability, data protection, access permissions, operating costs and human review will determine whether organizations trust agents with important business activities.

Looking Ahead

The integration gives Nous Research an opportunity to move Hermes beyond individual developers and AI enthusiasts into mainstream business workflows. Its combination of persistent working memory, reusable skills, shared organizational knowledge and Microsoft infrastructure could make it an interesting alternative to conventional workplace chatbots.

The next challenge will be demonstrating measurable productivity improvements in real business environments. As companies experiment with AI agents, the competition is increasingly shifting toward systems that do not simply generate answers but can complete work, retain useful knowledge and operate securely within existing enterprise software.


SEO Title: Nous Research Brings Hermes AI Agent to Microsoft Teams

Meta Description: Nous Research partners with Microsoft to bring Hermes Enterprise to Teams, allowing companies to delegate tasks, reuse AI skills and share knowledge.

Suggested URL Slug: nous-research-hermes-agent-microsoft-teams

Focus Keyword: Hermes Agent Microsoft Teams

Secondary Keywords: Nous Research, Hermes Enterprise, Microsoft AI agents, Collective Wisdom, Azure Container Apps Sandboxes, open-source AI agent

Get the day’s top stories in your inbox

One concise email. No spam, unsubscribe anytime.