Kisi MCP. Manage physical access, locks, and users via conversation.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Kisi MCP Server lets your AI agent automate physical access control. It connects to your cloud-based locks, doors, and user directories.
You can list all locks, check a lock's real-time status, manage user profiles, audit access groups, and trigger immediate remote unlocks—all from natural language.
It puts your facility's security status directly into your workflow.
What your AI agents can do
Get lock details
Retrieves full details for a specific lock, including its ID and current status.
Get my profile
Gets the profile details of the user currently authenticated to the system.
Get place details
Retrieves detailed information about a specific physical location or site.
Sends an unlock command to a specific door, regardless of its current status.
Retrieves the real-time operational status and details for one specific lock ID.
Lists all role assignments across the facility to see who has access to what.
Retrieves a list of all users managed within the Kisi organization, including profile details.
Lists all physical places (sites) configured in the Kisi environment, along with their details.
Lists all defined access groups used for managing permissions across sites.
Ask AI about this MCP
Supported MCP Clients
Waiting for input…
Kisi MCP Server: 9 Tools for Access Control Management
Use these tools to query lock status, manage user profiles, and perform facility access audits directly through your AI client.
019d75c1get lock details
Retrieves full details for a specific lock, including its ID and current status.
019d75c1get my profile
Gets the profile details of the user currently authenticated to the system.
019d75c1get place details
Retrieves detailed information about a specific physical location or site.
019d75c1list access groups
Lists all groups that define who can access specific areas.
019d75c1list locks
Lists every lock installed in your organization's managed locations.
019d75c1list places
Lists all physical sites or locations configured in your Kisi account.
019d75c1list role assignments
Lists all role assignments to audit permissions across the entire facility.
019d75c1list users
Lists all users in the organization to identify specific user IDs.
019d75c1unlock door
Sends a command to instantly unlock a specified door or lock.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Kisi, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Your AI agent can automate physical access control using the Kisi MCP Server. You can force a door to unlock with unlock_door and check a single lock's real-time status using get_lock_details. You'll find all installed locks by running list_locks, and you can pull full details for a specific lock with get_lock_details.
You can list every user in the organization with list_users and grab a specific user's profile with get_my_profile. You'll see all defined access groups using list_access_groups, and you can audit permissions across the whole site by calling list_role_assignments. To map out where you are, you can list all physical sites with list_places, and you'll get detailed info on a specific location using get_place_details.
You can also see all the role assignments that govern access using list_role_assignments.
How Kisi MCP Works
- 1 First, subscribe to the Kisi server and provide your Kisi API Key (Personal Access Token).
- 2 Your AI client authenticates using this token and gains access to your facility's live data.
- 3 You instruct your agent (e.g., 'Unlock the main office door.') and the agent uses the appropriate tool to execute the command.
The bottom line is, your AI agent uses your API key to talk directly to your Kisi system, letting you manage security tasks without opening a dashboard.
Who Is Kisi MCP For?
The facilities manager who has to manually check door statuses across multiple sites is the primary user. It's also critical for security teams needing instant, auditable access. Developers who need to integrate physical security status into custom apps also rely on this. This tool takes security operations out of the dashboard and into the chat window.
Uses the server to quickly check if a remote door is locked or unlocked, or to list all users to verify who has site access.
Triggers immediate remote unlocks for emergencies and runs reports on who has what permissions using list_role_assignments.
Integrates the list_locks and get_lock_details tools into custom applications to build custom status monitoring dashboards.
What Changes When You Connect
- Remote control of physical assets: You can use the
unlock_doortool to trigger an immediate unlock on any managed door, bypassing the need to open the main dashboard. - Real-time device visibility: The
list_locksandget_lock_detailstools let you list every lock and check its current status (online/offline, locked/unlocked) instantly. - Simplified user auditing: Instead of navigating multiple menus, you can run
list_role_assignmentsto query groups, places, and role assignments to monitor who has access. - Contextual location management: The
list_placesandget_place_detailstools help you understand the physical context, letting you know exactly which site a lock belongs to. - User and group control: You can use
list_usersandlist_access_groupsto quickly pull user IDs or see all defined access groups without leaving the chat. - Development speed: Developers gain access to direct API calls via tools like
get_my_profileandget_place_details, letting them embed security logic into custom applications.
Real-World Use Cases
Emergency door access during an incident
A security analyst needs to open a locked door immediately. Instead of opening the dashboard and finding the lock ID, they ask their agent to 'Unlock the Main Entrance door (ID: 12345)'. The agent runs unlock_door and confirms the action.
Auditing user access before an audit
A compliance officer needs to verify who has elevated access rights. They run list_role_assignments and list_users to cross-reference role assignments against the full user list, providing a clear, auditable report.
Troubleshooting a locked site
The facilities manager notices a door is acting strangely. They run list_locks to see all device IDs, then use get_lock_details on the specific lock to diagnose if it's offline or if the lock itself is reported as locked.
Onboarding a new location
A developer setting up a new branch needs to check the physical context. They use list_places to find the site ID, then run get_place_details to confirm the location's specific locks and required access groups.
The Tradeoffs
Assuming the AI knows the ID
Asking the agent, 'Unlock the main entrance door.' The agent fails because it needs a specific ID or name for the unlock_door tool.
→
First, run list_locks to get the list of device IDs. Then, instruct the agent: 'Using the ID from the list, run unlock_door for the Main Entrance.' This provides the necessary context.
Mixing up roles and users
The user runs list_users and thinks they see the permissions. This is wrong; list_users only shows profiles, not what those users can actually do.
→
To see permissions, you must run list_role_assignments. This tool shows the linkage between a user/group and a specific resource, providing the actual access control data you need.
Over-relying on manual web searches
Searching the Kisi dashboard for 'access group list' and manually copying out the names to give to a developer.
→
Just ask your agent to run list_access_groups. The output is clean, machine-readable data that you can immediately paste into code or a document.
When It Fits, When It Doesn't
Use this if your core pain point is managing physical security or access control across multiple, disparate locations. You need to turn operational data—like lock status, user roles, or facility location IDs—into actionable commands or reports without opening a dedicated dashboard. You're looking for automation that requires specific data points, like a lock ID or a place ID.
Don't use this if you only need general status information that doesn't involve credentials (e.g., 'What's the weather?'). Use a general information retrieval tool for that. Also, if you only need to view user details without checking permissions, list_users is sufficient, but if you need to know what they can do, you need list_role_assignments.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Kisi. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 9 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking facility status shouldn't require jumping through four different tabs.
Today, checking a simple status—like seeing if the storage room door is locked—means logging into the Kisi portal. You navigate to 'Devices,' find the lock ID, then you check the status. If you need to know who can access it, you leave that tab and go to 'Users & Roles.' It's a three-step, click-heavy process that wastes time and makes audits hard.
With the Kisi MCP Server, you just tell your agent, 'Check the status of the storage room door.' The agent runs `get_lock_details` and returns the current status instantly, giving you the data without touching a web browser. You get immediate, actionable status reports.
Unlock Door MCP Server: Remote control from chat.
Before, if an employee was locked out or needed immediate access, a physical key or a manual command had to be sent through a specific interface. This required knowing the exact door ID and navigating to the unlock function.
Now, you simply tell your agent to 'Unlock the main entrance door.' The agent executes `unlock_door`, sending the command directly to the lock. It's instant, documented, and works regardless of your current desktop workflow.
Common Questions About Kisi MCP
How do I find the ID for the Main Entrance door using the Kisi MCP Server? +
Run the list_locks tool. This returns a comprehensive list of all locks, giving you the necessary unique IDs to reference in other commands, like unlock_door.
What is the difference between `list_users` and `list_role_assignments` in the Kisi MCP Server? +
list_users provides basic profile information (names, emails). list_role_assignments tells you what those users are actually allowed to do—it’s the permission audit tool.
Can I use the Kisi MCP Server to check if a place is active? +
Run list_places first. This lists all configured sites. Then, use get_place_details on the specific place ID to confirm its current details and associated locks.
Does `get_lock_details` require a specific lock ID? +
Yes, it does. You must provide the specific lock ID when running get_lock_details to retrieve the correct, targeted information.
What if I need to unlock a door that isn't listed in the system? +
The system will fail the unlock_door command. This means the door ID you provided is invalid or the lock is not managed by Kisi. You need to verify the ID using list_locks.
How do I use `list_locks` to check the online/offline status of my doors? +
The list_locks tool returns a real-time status for every managed lock. This status tells you if the device is connected (online) or if it's reporting an issue (offline).
What data does `get_place_details` provide for a specific location? +
It provides comprehensive details about a location, including associated active locks and the number of access groups linked to that physical place. This helps map out site configurations.
What happens if I try to `unlock_door` with an invalid lock ID? +
The system throws a specific error indicating the lock ID is invalid or inaccessible. Your agent can catch this error and prompt you to check the ID or confirm the door status.
How do I perform a remote unlock using the agent? +
Use the unlock_door tool with the unique ID of the lock you wish to open. Make sure the authenticated user has the necessary permissions for that door.
Can I see if a door is currently online? +
Yes, the list_locks and get_lock_details tools return the real-time connectivity status (online/offline) for every device.
Is it possible to list all users in my organization? +
Absolutely. Use the list_users tool to retrieve a comprehensive list of all members registered in your Kisi organization.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
IPGeolocation.io
Real-time IP geolocation, security intelligence, and timezone data — identify locations, detect VPNs, and manage timezones directly via AI.
Runlayer
AI enterprise control plane: manage MCP servers, skills, agents, and security policies via agents.
AbuseIPDB
Audit IP addresses — check abuse scores and reports via AI.
You might also like
LawPay
Accept legal payments online with trust and operating account compliance that meets bar association requirements for law firms.
Gainsight CS
Manage customer success, track health scores, and oversee the timeline via AI agents with Gainsight CS.
ThriveCart
Manage your ThriveCart sales and products — audit transactions and subscriptions via AI.