Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsSome links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
To host a remote MCP server, implement it with an official MCP SDK or FastMCP, expose the current Streamable HTTP transport on one HTTPS endpoint that accepts POST requests, deploy it to an HTTP-capable platform such as Google Cloud Run, and protect it with authentication and Origin validation. Keep legacy HTTP+SSE compatibility only if clients you need to support require it.
Remote hosting means the MCP server runs on service infrastructure and clients reach it over HTTP; that differs from a local server launched over standard input/output (stdio). Cloud Run supports remote Streamable HTTP and SSE servers, but not stdio servers. This guide focuses on Cloud Run because its documentation directly covers remote MCP deployment; check another provider’s current transport, streaming, and authentication support before choosing it.
Choose the transport before choosing a host
For a new remote service, use Streamable HTTP. The MCP specification describes it as the replacement for the older HTTP+SSE transport. In the revision dated 2026-07-28, the specification removed the standalone GET stream and protocol-level session behavior, making it important to check that your client and server agree on the transport revision they support.
Recommended Free Tools
| Transport or mode | How it connects | When to use it |
|---|---|---|
| Streamable HTTP | One MCP endpoint accepts POST requests. A response can be a JSON object or a request-scoped SSE stream. | Recommended for a new remote server. |
| Legacy HTTP+SSE | Older remote transport with a separate event-stream pattern. | Retain only when a required older client depends on it; verify compatibility against that client’s supported revision. |
| stdio | A local client communicates with a server process using standard input and output. | Local integrations, not Cloud Run remote hosting. |
For Streamable HTTP, expose a single endpoint path such as https://example.com/mcp. Each JSON-RPC request or notification is sent as its own POST. The response may be JSON or an SSE stream carrying request-scoped notifications and the final response. Do not design a new implementation around an independent GET stream or protocol-level sessions if you are targeting the specification revision dated 2026-07-28.
#1 Best Overall
- 【Powerful Load-bearing】12U Network Rack Open Frame is constructed from durable cold rolled steel; Rack shelf supports enhance stability, wall-mounted capacity of 130lbs, the ground-mounted up to 260lbs
- 【Considerate Designs】Open-frame layout, including a top panel adding space, anti-slip shelf stops fixing devices and compatible racks for stack and expansion to meet requirements of home server rack
- 【Complete Accessories】A 12U open frame server rack, two ventilated shelves, four shelf stops, four velcro straps and a set of equipment mounting screws
- 【Versatile Application】Ideal for space-efficient multi-device setups in warehouses, retail, classrooms, offices and more; Excellent choices as AV Rack/IT Rack
- 【Effortless Setup】 Network Rack includes hardware, a comprehensive manual, mounting hole drilling template and an online assembly video to simplify setup
Build a server that speaks remote MCP
Use an official MCP language SDK or FastMCP rather than implementing the protocol yourself. Google recommends either for new servers, and the official TypeScript SDK identifies Streamable HTTP as its recommended transport for remote servers. The SDK should handle protocol details; your application supplies the tools or other capabilities you intend to expose.
Before deployment, confirm that the server is configured for Streamable HTTP, serves the one POST-capable MCP endpoint, and can return the response forms your clients support. If you need older-client support, the TypeScript SDK documents a compatibility server for old and new clients; check its current API and the target client’s transport support rather than assuming compatibility from the server alone.
The available deployment guidance does not prescribe one language, SDK version, tool definition, or application-specific server implementation. Use the current API documentation for your selected SDK to write that part of the program, and pin and test the SDK version you deploy. The concrete Cloud Run deployment commands below apply once you have a valid source tree or container image for your server.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Deploy the endpoint on Cloud Run
Cloud Run supports MCP servers using Streamable HTTP or remote SSE, can deploy from source or a container image, provides an HTTPS URL, and supports HTTP response streaming. A stdio-only process is not a remote MCP service for Cloud Run. In a container, configure the process to listen on the port supplied by the platform.
Rank #2
- ADJUSTABLE DEPTH: 4-Post 42U open frame server rack with 4 vertical rails and adjustable mounting depth 22" to 40" (56,0cm to 101,7cm); Compatible with various servers / switches / data / AV and other IT equipment; EIA/ECA-310-E Compliant
- EASY ASSEMBLY: Mobile network rack with easy-to-follow assembly instructions and online video; Compact flat-pack shipping to avoid damage and facilitate installation; Total product height of 80.3in (204 cm) with casters, 78in (198cm) without casters
- COLD ROLLED STEEL: Durable 4 Post 19in open frame rack designed for ventilation with 42U mounting height and 1320lb (600kg) weight capacity (stationary); 3 install options included: casters, levelling feet, or base-plate to secure rack to the floor
- HARDWARE INCLUDED: Rolling computer/data rack includes cage nuts and screws to mount equipment, easy to read Units (U) and depth adjustment markings, cable management hooks for organization, and required assembly tools
- THE IT PRO'S CHOICE: Designed and built for IT Professionals, this 42U rack is backed for 2-years, including free lifetime 24/5 multi-lingual technical assistance
- Prepare the service. Configure the SDK server for Streamable HTTP and set its MCP endpoint path, for example
/mcp. Make the process listen on the platform-provided port. Decide whether Cloud Run IAM or application-level authentication will control access before making the endpoint available. - Deploy from a source tree. From the directory containing the deployable source, run
gcloud run deploy --source .. Follow the deployment prompts or supply the service configuration appropriate to your project. - Or deploy a container image. Build and publish your server image, then run
gcloud run deploy --image IMAGE_URL --port PORT, replacingIMAGE_URLwith the image reference andPORTwith the port your container listens on. - Record the service URL and endpoint path. Cloud Run supplies an HTTPS URL. Combine it with the configured MCP path, such as
/mcp, and give that complete endpoint to clients. If the service requires authentication, client requests must follow the chosen identity flow. - Verify the transport from a real client. Send a supported MCP request to the endpoint and confirm that the client receives a valid JSON or request-scoped SSE response. A successful deployment alone does not prove that the endpoint path, transport revision, and client are compatible.
Cloud Run supports HTTP response streaming for this use case. That is useful when a Streamable HTTP response is delivered as SSE, but it does not remove the need to check the client’s transport behavior or the SDK’s current implementation details.
Choose authentication based on where the client runs
Do not treat an HTTPS URL as a security boundary by itself. The MCP Streamable HTTP specification requires servers to validate the Origin header on incoming connections and return HTTP 403 for an invalid Origin. It also recommends proper authentication for all connections. Origin validation helps address DNS rebinding; authentication restricts who can invoke the service.
Local client connecting to Cloud Run
For a local MCP client, Cloud Run’s default IAM Invoker policy can protect the service. Google documents two approaches: run gcloud run services proxy locally so the proxy injects the operator’s identity, or send an OIDC ID token whose audience matches the service URL. Choose the approach that fits the client’s configuration and identity setup. A client that cannot supply the required identity will not be able to invoke an IAM-protected service directly.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Client and server both run on Cloud Run
For Cloud Run-to-Cloud Run use, Google documents a sidecar for communication within the same instance, service-to-service authentication for separate services, or Cloud Service Mesh for managed authentication and traffic controls. These are different deployment patterns, not interchangeable endpoint settings: choose based on whether the client and MCP server share an instance or run as separate services.
Rank #3
- Adjustable Depth: 23-40'' adjustable depth is used for servers and network equipment, ensuring enough space for AV equipment, components, and cabling, while allowing you to access ports and equipment from multiple sides.
- Strong Load Capacity: Ground-Mounted Load Capacity: 500 lbs, Wall-Mounted Load Capacity: 150 lbs. The av rack is made of carbon steel for better weldability performance and can help save space while meeting your need to place multiple devices.
- User-friendly Design: Ergonomic design makes the open frame av rack easier to use. The additional top panel is able to place other items with more available space. Roller design moves anywhere and anytime, is convenient, and is more energy-saving.
- Complete Accessories: We provide the accessories you need, including 2 x Pallets, 145 x M5*10 Cross Head Screws, 4 x Casters, 4 x M10*50 Expansion Screws,10 x M6*12 Cage Nuts, 1 x Grounding Wire, 1 x User Manual.
- Wide Application: The server rack wall mount maximizes the use of available space, suitable for retail venues, classrooms, offices, and other places where space is limited.
Local server versus remotely hosted server
The specification says local servers should bind to 127.0.0.1 and recommends authentication for all connections. Those local-binding recommendations should not be confused with the network binding needed for a container platform to reach a remotely hosted process. For Cloud Run, follow the platform’s supplied-port deployment requirement and rely on the service’s authentication and Origin checks to protect the remote endpoint.
Test the endpoint and plan for operational limits
Check the service from the same kind of client and network path that will use it. Verify all of the following before relying on it:
- The configured endpoint path is the one clients are given, and it accepts POST requests.
- The SDK and client support the same transport behavior, including the relevant Streamable HTTP revision.
- Responses work when delivered as JSON and, where used, request-scoped SSE.
- Invalid Origin values are rejected with HTTP 403, and unauthenticated requests do not gain access.
- The selected identity flow works from the client’s actual location: local machine, same Cloud Run instance, or another service.
There is no performance target or cost figure established here for a particular workload. Measure your own request latency, response duration, concurrency behavior, and platform charges under representative use, and check current Cloud Run pricing and regional availability before committing to a region or budget. Geographic placement should account for where clients and any services the MCP server calls are located. Capacity, scaling, observability settings, and total cost depend on your service configuration and workload; they are not guaranteed merely by choosing Cloud Run.
Troubleshoot common remote MCP failures
Client reports that the endpoint or method is not supported
Check that the client is using Streamable HTTP rather than stdio or legacy HTTP+SSE, and that its expected MCP revision matches the server. Confirm that the URL includes the exact configured MCP path and that the endpoint accepts POST. A client built around the older separate-stream behavior may need a compatibility implementation or an updated transport.
Rank #4
- Universal 19” Rack Mount Compatibility – Perfect for pro audio, video, IT, and network gear. Compatible with mixers, routers, patch panels, servers, power amps, and more.
- Heavy-Duty Load Capacity – Built to support up to 550 lbs. Ideal for studio gear, DJ setups, server equipment, and AV components that demand serious stability.
- Robust Steel Frame & Design – Made with 1.5mm thick steel and weighs 36 lbs for maximum durability, reduced vibration, and long-term reliability in any setting.
- Mobile & Secure – Preinstalled with 3” industrial-grade caster wheels (lockable), making it easy to move and position your rack exactly where you need it.
- All-In-One Setup Kit Included – Comes with 34 rack screws (5mm & 6mm), a 1U blank spacer, and an assembly tool—ready for fast installation out of the box.
Requests receive HTTP 403
A 403 can result from an invalid Origin, an authentication or IAM policy that denies the caller, or a missing or mismatched OIDC identity. Check the Origin validation rule and the caller’s identity separately; do not disable Origin validation as a shortcut. For the documented OIDC approach, ensure the token audience matches the Cloud Run service URL.
The service deploys but cannot be reached
For a container deployment, verify that the server listens on the platform-provided port and that the deployment’s --port value matches it. Confirm that the MCP path is part of the URL passed to the client. A process that only exposes stdio is not a remotely reachable HTTP MCP endpoint.
Streaming responses stall or fail in a client
Cloud Run supports HTTP response streaming, but client and SDK behavior still matters. Confirm the client accepts the server’s response type, test a request that produces the relevant SSE response, and verify that the selected SDK is configured for Streamable HTTP. If only older clients fail, investigate transport compatibility rather than assuming the host cannot stream.
Local requests work but service-to-service calls fail
Recheck which topology you deployed. Same-instance communication, separate Cloud Run services, and a local operator connecting to Cloud Run use different documented identity patterns. Select the corresponding sidecar, service-to-service authentication, or local proxy/OIDC method and verify the identity at the calling service.
Best Value
- Adjustable Depth: Depth adjustable from 23" to 40", this open frame server rack accommodates servers and network equipment while providing ample space for A/V gears and cable management. Enjoy easy access to ports and devices from multiple angles.
- High Weight Capacity: Supports up to 300 lbs on the floor (200 lbs when adjusted to maximum depth) and 200 lbs when wall-mounted (depth cannot be adjusted in wall-mounted mode). Made from carbon steel for superior welding performance and durability, this open frame rack is designed to save space while accommodating multiple devices.
- User-Friendly Design: Designed with your convenience in mind, this open frame server rack features an top shelf for extra storage and improved space utilization. The rolling casters let you move it effortlessly wherever you need it, making setup and movement a breeze.
- Widely Applicable: Maximize your space with this adaptable open frame server rack, designed to make the most of every inch. Ideal for retail spots, classrooms, offices, and any area where space is at a premium, it delivers practical solutions for your storage needs.
- Everything You Need: Our open-frame rack comes with fully equipped accessory kit for easy setup and secure installation: 2 x Trays, 4 x Casters, 1 x set of Screws, 16 x M6*12 Cage Nuts, 1 x Grounding Wire, 1 x Internal & External Hex Wrenches, and 1 x User Manual.
Or skip the browser setup
ScreenshotNeo is a separate website screenshot API and MCP server, not a host for the MCP server you are deploying in this guide. If your agent also needs website screenshots, one GET request can return an image or PDF. See the ScreenshotNeo API documentation.
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpScreenshotNeo accepts cookie and consent banners like a visitor and removes more than 60 known consent platforms, newsletter popups, and chat widgets before capture; each step can be turned off. Bot checks and other clean-shot failures—including blank pages, timeouts, and failed loads—are not billed, and responses identify the page verdict and billing status. It also provides an MCP server with screenshot, page-info, and PDF-capture tools for AI agents. The free plan includes 1,000 shots per month with no card; paid plans start at $5 for 3,000 shots.
Sign up for 1,000 free ScreenshotNeo screenshots a month, with no card required.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →To compare hosting choices for your own MCP service, evaluate transport support, HTTPS and streaming, authentication, source or container deployment, observability and scaling, geographic placement, client compatibility, and total cost. Cloud Run is a documented option for remote MCP; verify the current details for any other provider before deploying.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
