Laravel Forge MCP. Run deployments and audit servers from chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Laravel Forge MCP Server manages your entire server ecosystem. Your AI agent connects to this server to list connected drops, inspect server structures, and run deployments.
It lets you query databases linked to specific domains and check worker configurations, all without touching the UI or running shell scripts directly.
It's full devops control, executed natively via chat.
What your AI agents can do
Deploy site
Runs a deployment script queue for a specific repository site.
Get server
Retrieves detailed information about a single connected server droplet.
Get site
Looks up specific details for a given site layout on a server.
List all connected Forge servers and retrieve detailed data on specific server droplets.
Run deployment scripts and monitor the output for a specific website or site layout.
List and query all active database clusters linked to domains managed by Forge.
Retrieve the configurations and status of queue workers running on a tracked site.
Get a list of active physical access keys inserted on the root server.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Laravel Forge MCP Server: 9 Tools for Server Ops
This server gives your AI agent full control over your Laravel Forge environment. Use these tools to deploy sites, audit servers, and manage databases from a single chat interface.
019d75c4deploy site
Runs a deployment script queue for a specific repository site.
019d75c4get server
Retrieves detailed information about a single connected server droplet.
019d75c4get site
Looks up specific details for a given site layout on a server.
019d75c4list databases
Lists all active database clusters mounted on the Forge server.
019d75c4list recipes
Retrieves available custom shell recipes within your team limits.
019d75c4list servers
Gets the master list of every connected Forge server.
019d75c4list sites
Lists all websites mounted to a specific server.
019d75c4list ssh keys
Retrieves active physical access keys inserted on the root server.
019d75c4list workers
Gets the queue worker configurations running on a specific tracked site.
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 Laravel Forge, 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 connects to this server to manage your whole server stack. You get full devops control without touching the UI or running shell scripts directly. You'll use the agent to list connected drops, inspect server structures, run deployments, query databases linked to domains, and check worker configurations. It's full control, executed natively in chat.
View entire server infrastructure
- You can use
list_serversto get the master list of every connected Forge server. You can then useget_serverto pull detailed data on any single server droplet.list_siteslets you see all websites mounted to a specific server, andget_sitegives you specific details for a given site layout. You can also runlist_ssh_keysto retrieve active physical access keys inserted on the root server.
Manage site deployments
- You'll use
deploy_siteto run a deployment script queue for a specific repository site. This lets you monitor the output while you're chatting.
Audit databases
- You can use
list_databasesto list all active database clusters mounted on the Forge server. You can query these clusters to check domains linked to them.
Check background workers
- You'll use
list_workersto get the queue worker configurations running on a specific tracked site.
How it works
- You subscribe to this endpoint and give your core Laravel Forge API token. Your agent then uses the necessary API calls to manage your entire environment.
How Laravel Forge MCP Works
- 1 Subscribe to the endpoint and pass your Laravel Forge API token.
- 2 Instruct your AI agent to perform a devops task (e.g., 'List all servers').
- 3 The agent executes the required tool calls and presents the results in chat.
The bottom line is, you give your agent your API key once, and it handles all the API calls for you.
Who Is Laravel Forge MCP For?
This is for the Ops Engineer who spends too much time clicking through dashboards just to check a server status. It's for the Lead Developer who needs to verify worker key metadata without context-switching. If your job involves running deployments or auditing network services, you need this.
Runs deployments natively through chat commands during sprint reviews or incident response.
Checks worker configurations and SSH key metadata for specific environments instantly.
Audits entire physical networks and deploys changes without running shell scripts manually.
What Changes When You Connect
- See the full server picture instantly. Use
list_serversto pull a master list of every connected Forge server without navigating through the UI. - Keep deployments in context. Run
deploy_siteand watch the deployment script queue execute and report status directly in your conversation. - Audit credentials easily. Use
list_ssh_keysto retrieve all active physical access keys attached to the root server, eliminating manual key searches. - Check background processes. Use
list_workersto confirm queue worker configurations are running correctly on a site, letting you verify job routing status. - Manage databases from chat. Run
list_databasesto query and see all active database clusters linked to your domains, skipping the database management page. - Inspect site details. Use
get_siteorlist_sitesto get the specific configuration details for a site on a server, helping pinpoint deployment issues fast.
Real-World Use Cases
Server-wide audit before scaling
A system admin needs to audit the entire network before scaling up. Instead of checking multiple dashboards, they ask their agent to run list_servers to get all droplets, then use list_sites on each one to verify which sites are active. This gives them a complete inventory in one chat session.
Debugging a failed deployment
A lead developer notices a site is down. They ask their agent to run list_servers to confirm the droplet ID, then use get_site to check the specific site configuration. Finally, they run deploy_site to push the latest commits and see the output immediately.
Verifying worker health after code changes
The ops engineer pushes a new feature. They ask the agent to run list_workers for the production site. If the workers show errors, they can then check the server's status using get_server to see if the underlying hardware is the problem.
Onboarding a new team member
A new team member needs to know what credentials are active. They ask the agent to run list_ssh_keys and list_recipes to see what physical access keys and custom shell recipes are available across the environment.
The Tradeoffs
Assuming the agent knows the scope
The user types: 'Deploy the site.' The agent asks for a site ID, and the user forgets to provide it, resulting in a failed, generic error message. The user then tries to guess the correct parameters manually.
→
Always specify the target. Start by asking the agent to run list_servers to get the master list, then list_sites to identify the exact site name, and finally, run deploy_site with the confirmed site ID.
Trying to run system commands directly
The user attempts to use the chat to run raw shell commands like ssh user@host 'command'. These commands fail because the agent is confined to the Forge API tools.
→
Use the specialized tools. To check the server status, run get_server. To check worker status, run list_workers. Let the agent handle the API calls.
Overlooking the required prerequisites
The user tries to run list_databases for a domain that hasn't been linked to Forge yet. The call fails, and the user doesn't know if the issue is the tool or the setup.
→
Verify the connection first. Use list_sites on the server first. If the site exists, then run list_databases to confirm the database link is active.
When It Fits, When It Doesn't
Use this if you need to manage multiple services across a complex, interconnected environment. Specifically, if you need to run a full lifecycle audit (e.g., Server -> Site -> DB -> Workers), this tool is necessary. You need the control that only a direct API connection provides, bypassing UI limitations.
Don't use this if you only need to view a single piece of data (like just a server's IP address). In that case, a simple API lookup might suffice. Also, don't use it if your goal is simply to write code—it's for operations, not development. For database-only tasks, a dedicated database client is better, but if the database is hosted by Forge, use list_databases.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Laravel Forge. 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
Auditing a server's current state shouldn't require five different tabs and three separate logins.
Right now, checking a server's full state is a pain. You jump into the main dashboard, click 'Servers,' then check the status. Next, you open the 'Sites' tab to see which websites are active, then you switch to 'Databases' to check the cluster status. You'll spend fifteen minutes just gathering a list of IDs and statuses, copying them into a spreadsheet.
With the Laravel Forge MCP Server, you just ask your agent. It runs `list_servers` and `list_sites` in sequence, compiling the entire operational status into a single, clean response. You get the full picture in seconds, not minutes.
Laravel Forge MCP Server: Deploy site updates from chat.
Before, deploying a site meant clicking the deployment button, waiting for the process to start, then refreshing the logs page repeatedly to see if it failed. If it failed, you had to open a whole new panel to figure out why.
Now, you tell your agent to run `deploy_site`. The agent executes the script and streams the full output back to you in chat. You see the success status, the error messages, and the validation flags—all in one place. It's immediate, and it's actionable.
Common Questions About Laravel Forge MCP
How do I check all connected servers using list_servers? +
You ask the agent to run list_servers. It returns a master list of all droplet IDs and their basic statuses. If you need deep details on one, you then follow up by asking it to run get_server on that specific ID.
What is the difference between list_sites and get_site? +
list_sites gives you a list of all websites mounted to a given server ID. get_site lets you drill down and retrieve the specific configuration details for one exact site layout.
Can I check worker status with list_workers? +
Yes, running list_workers retrieves the queue worker configurations for a site. It tells you if the daemon is running and what payload types it's handling, which is key for job processing.
Do I need to provide an API token for deploy_site? +
Yes, the agent requires your Laravel Forge API token to run deploy_site. This token grants the agent the necessary permissions to execute deployment scripts on your behalf.
How do I list databases using list_databases? +
Simply ask the agent to use list_databases. It returns a list of all active database clusters associated with the Forge server, allowing you to see what's connected.
How can I use list_ssh_keys to manage server access? +
list_ssh_keys retrieves all physical access keys tied to the root server. You use this to audit who has access credentials and confirm key existence before granting or revoking access.
What is the purpose of the list_recipes tool? +
list_recipes retrieves available custom shell scripts within your team's limits. This shows you what operational automation scripts are available to run on the server.
Does deploy_site handle all types of deployment errors? +
deploy_site executes the deployment script queue on a repository site. It captures the output structure, allowing your agent to diagnose specific failures during the build process.
How do I get started? +
Subscribe, copy your active API Token (which you generate inside Forge Dashboard → API), and connect it. The integration expects standard connections so you can trigger real build routines natively without intricate pipelines or ssh logic.
Can my AI automatically trigger site deployments? +
Yes! Give the agent the Server ID and Site ID you're questioning. It utilizes deploy_site to physically POST to Forge's pipelines kicking off an automated release. Perfect for hands-free releases right from Cursor.
What happens when I want to fetch metrics on queued application workers? +
Provide the exact IDs. The tool evaluates endpoint rules (list_workers) and unpacks default configurations running PHP queues natively back to the terminal screen.
Does it list physical server setups efficiently? +
Absolutely. Execute 'list_servers' or probe 'list_ssh_keys'. It aggregates root connection statuses, IPs, and structural records giving immense monitoring peace-of-mind at the keyboard.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Coding.net
All-in-one DevOps platform — manage projects, repositories, and issues via AI.
Portainer
Manage Docker containers and environments via Portainer — list, create, and start containers directly from your AI agent.
Azure DevOps
Manage work items, track builds, and coordinate releases across your Azure DevOps organization with full pipeline visibility.
You might also like
Hanko (Passkey Auth)
Manage passkey authentication and user identity via Hanko — initialize registrations, handle logins, and manage WebAuthn credentials directly.
Toggl Plan
Manage your team's visual timelines, track project phases, and balance workloads securely via your AI agent.
WordPress Subscriber Creator
This MCP does exactly one thing: it registers a new user in your WordPress database with the strictly enforced 'Subscriber' role. Perfect for turning Claude into an automated lead generation or membership onboarding bot.