Use cases

Where teams start with Ospiri.

The situations security and IT teams bring to us most often, and what Ospiri does about each one.

Use case · Large enterprise on Microsoft

Copilot is going out company-wide. Leadership wants one view of every agent.

Who this is for

VP of Technology and VP of Infrastructure.

The problem

The company has standardized on Microsoft. Copilot is rolling out to every team, and engineering builds its own agents too. Each tool has its own console and its own logs. Two plain questions have no single answer: which agents are running, and what data are they touching?

What Ospiri does

  • Starts with a one-week assessment: a lightweight scanner lists the agents, AI apps, skills and MCP servers on the endpoints, including the ones built in-house.
  • Joins what it finds by agent and by user, so Copilot and the company's own agents sit in one record.
  • Shows what each agent can reach: files, credentials and internal systems.
  • Adds limits where they matter, one group of agents at a time.

What they get

  • Technology leadership: one view of every agent in use, approved or not.
  • Infrastructure: who uses Copilot and other agents, on what data, and where data is exposed.
Use case · Financial services

Agent rollout was on hold. It is back on, with access set by team.

Who this is for

CISO and Head of IT.

The problem

The lender has to protect customer data: borrower tax returns, bank statements, credit memos and payroll exports. An agent runs with everything its user can open, and years of file permissions put most of that within reach. So security has stopped employees from using agents. The clean fix is an IAM modernization project, and nobody wants to wait for it.

What Ospiri does

  • Sets where each team's agents can read from and write to. Underwriting agents read the loan files in their own pipeline. They cannot open HR, treasury or another team's borrowers.
  • Enforces it inline on the endpoint, at the moment an agent opens or changes a file. The employee's own access does not change.
  • Uses the identity groups the company already has. No shares are re-permissioned.
  • Sends an out-of-bounds write to a safe copy. The work continues and the loan record is untouched.
  • Keeps a timeline of what each agent read and wrote, ready for audit and exam requests.

What they get

Agents are back on for underwriting, servicing and finance, each with limits the CISO can show an auditor. The IAM project runs on its own schedule.

Use case · Software company

Block the note-takers and app builders you have not approved. Keep the ones you have.

Who this is for

Head of Security and VP of Engineering.

The problem

People are installing AI tools faster than security can review them. Note-taking apps record customer calls and roadmap meetings, often on a personal account. Some run on the laptop and never join the call as a bot, so nobody sees them. Vibe-coding tools generate an app, connect to the repo and deploy to outside hosting in an afternoon. A blanket ban gets ignored.

What Ospiri does

  • Lists every AI note-taker and app builder on every endpoint, running or dormant, and shows whether it is signed in with a personal or a company account.
  • Blocks unapproved note-takers from starting or recording. The approved note-taker keeps working for the teams allowed to use it.
  • Blocks app builders by type. For example: tools that push source code to outside hosting, or tools on personal accounts. Approved coding agents keep running.
  • Holds a new tool for approval, so the employee knows what to ask for.

What they get

An approved list that is enforced on the endpoint, and a running view of what people tried to use. That view tells security what to approve next.

Do you know what AI is running on your endpoints now?

A lightweight scanner lists the agents, skills and MCP servers on the endpoints you choose. We walk through the findings with you, and the scanner is removed when the week ends.

Add control later, on the machines where it matters.