> Instead, I did something a little bit hacky: if the request is for GET /mcp and the Accept header includes text/html and NOT application/json or text/event-stream, I return a HTML page explaining to the user they're trying to view an mcp server and they need to add it to their client.
This feels like less of a hack and more of discovering what some of the HTTP headers are for. You’re choosing rather reasonably how to present the resource found at /mcp when a client is asking for the resource to be presented in HTML format. It’s perfectly fine to offer an HTML response that says “hey this is not really presentable in HTML. Do this instead.”
I love this. The best documentation is that which is presented at the exact place and time that it is needed, and users self support really well if you do that.
I once worked for $COMPANY and we had a network scanning application. Always generated a lot of tickets from angry people wanting to scream about bots.
So we put a web page page on each worker that would inherit some details from whatever job was running, and say “I’m a $TYPE_OF_SCANNER FROM $COMPANY doing $THING_THAT_BENEFITS_YOU.
This behavior is covered by our terms of service page at $LINK.
If you believe that we should not be doing this, please contact $SUPPORT and provide this code:
Doing a workshop this week on MCP for an enterprise client and explaining the 406 returned by GET against /mcp w/o text/event-stream is exactly one of the things that I have to bring up when I do these.
The specification still leaves a lot to be desired, especially as it relates to auth. There are lots of bad ways to do auth with MCP and only a couple of good ways. It also puts a lot of pressure on the various IdP vendors and relies on lesser used areas of OAuth 2.0/2.1 (like DCR, token exchange, etc.). It started out in a place where the assumption was you were running an MCP Server on a laptop or you were a SaaS provider serving lots of individual users -- somehow DCR in the initial spec iteration seemed like a good idea (spoiler: it wasn't) and fortunately, the latest revision has somewhat addressed that. XAA/ID-JAG & CIMD should continue to round-out some client management and auth solutions for the enterprise.
Gateways are another area that needs to be addressed in the spec. There isn't a formal definition of one in the spec, and yet, there are lots of "gateways" out there. What a gateway is and what it should do is an open question and it means different things to different people depending on who you ask. For example, who does token exchange: the MCP server or the MCP gateway? Both are valid answers right now depending on the implementation or your opinion of which is best.
More spec iterations should be coming this year. I'm still pretty optimistic about MCP as a whole, as it remains a good way to standardize tool calls across agents and some of the other entities that it provides like resources and prompts are genuinely useful to add more determinism to an agent. Interceptors and skills will be good, too.
If you're interested in helping to evolve the spec further, the MCP Contributors Discord is active. There are lots of IGs/WGs that solicit feedback and you can participate in meetings with your feedback.
I have also been finding the MCP auth story to be really lacking was excited to see OAuth 2 support until I tried to get it to work at work and realized our idp implementation didn't support 2.1, and went into the spec and started wondering if anyone had a good experience yet. Luckily most of our environment can settle on a OAuth token env var standard until that's all in order.
A lot of how well it works or won't work depends on your clients, as not all clients have support for things like RFC 9728 (Protected Resource Metadata). Assuming your client has good support for most of the OAuth 2.0 standards that MCP uses (you don't need DCR as you can statically register clients, assuming that is viable for your environment), then it is possible now with most IDPs to get an OAuth 2.0 auth code flow working just fine. You would then do a token exchange to the upstream to ensure to get the appropriate new audience and rescope/downscope as necessary. Gateways can also help here a lot as instead of baking in all of the auth concerns into your MCP Servers and upstreams, you delegate that to the MCP Gateway. Again, gateway here means different things to different vendors, but Kong, for example, has the ability to act as an MCP proxy (gateway), expose tools based on consumer role or group, apply OAuth 2.0 to it and do an upstream token exchange, while also acting as a regular API gateway that can protect an endpoint with OAuth 2.0/OIDC.
Excellent. Tiny affordances like this should be more commonplace, IMO.
I think this probably also helps when truly clueless users drop the link into an agent directly, because then the agent will relay the message to the user.
> I did something a little bit hacky: if the request is for GET /mcp and the Accept header includes text/html and NOT application/json or text/event-stream, I return a HTML page
I think so, but if you wanted to be pedantic you could make the argument that the same resource is not being served with the different Accept header. A welcome page is not the same thing as the JSON returned from the MCP spec.
A redirect actually does make sense but then there will be some percentage of users that end up copying the redirect URL and try to use that as the MCP URL
I never quite understand why /mcp endpoint is needed.
You can still keep using Rest API with swagger docs and tell the AI to read the swagger docs. It's the same thing. The entire Rest API specification is a lot more flexible than the JSON RPC format /mcp uses.
> Despite the fact that MCP is an utterly terrible attempt at a "specification",
Can I just say that anybody involved with MCP's launch should be ashamed of what they put out there. I understand tool calling. I understand specs. I read MCP's "spec" and I used useless word salad that alternates between baby's first wire format and pie-in-the-sky marketing speak. Several of the navigation links I encountered were broken.
Poorly thought out, poorly communicated, but it's the only thing out there that 1) meets the need, and 2) published by people with a huge amount of reach. Right place, right time, shit effort. So it gets adoption. Like the history of PCs, of the internet, or everything, I guess. Worse truly is better.
But why would you show the MCP server URL as a clickable link in the first place, if it's not meant to be clicked? Put it in a monospaced box with a "copy to clipboard" button, it's not the fault of the user for not "thinking ahead" when they click a clickable link that wasn't actually meant to be clicked.
I call this "dev think". Blames user for not thinking ahead when they themselves haven't thought ahead about how users use things. It's everywhere these days, as an end user of things myself, it's maddening.
> I wish the spec had some capability to mitigate for this
I like their solution. It feels like because remote MCP servers are built on HTTP that they are actually using the spec as intended. Serving html when asked for it.
> Instead, I did something a little bit hacky: if the request is for GET /mcp and the Accept header includes text/html and NOT application/json or text/event-stream, I return a HTML page explaining to the user they're trying to view an mcp server and they need to add it to their client.
This feels like less of a hack and more of discovering what some of the HTTP headers are for. You’re choosing rather reasonably how to present the resource found at /mcp when a client is asking for the resource to be presented in HTML format. It’s perfectly fine to offer an HTML response that says “hey this is not really presentable in HTML. Do this instead.”
Indeed! Far from a hack it is the entire purpose to the accept header!
I thought the same thing, just discovered content negotiation!
I love this. The best documentation is that which is presented at the exact place and time that it is needed, and users self support really well if you do that.
I once worked for $COMPANY and we had a network scanning application. Always generated a lot of tickets from angry people wanting to scream about bots.
So we put a web page page on each worker that would inherit some details from whatever job was running, and say “I’m a $TYPE_OF_SCANNER FROM $COMPANY doing $THING_THAT_BENEFITS_YOU.
This behavior is covered by our terms of service page at $LINK.
If you believe that we should not be doing this, please contact $SUPPORT and provide this code:
$SCAN_JOB_IDENTIFIER”
Call volume and unhappy customers went way down.
Doing a workshop this week on MCP for an enterprise client and explaining the 406 returned by GET against /mcp w/o text/event-stream is exactly one of the things that I have to bring up when I do these.
The specification still leaves a lot to be desired, especially as it relates to auth. There are lots of bad ways to do auth with MCP and only a couple of good ways. It also puts a lot of pressure on the various IdP vendors and relies on lesser used areas of OAuth 2.0/2.1 (like DCR, token exchange, etc.). It started out in a place where the assumption was you were running an MCP Server on a laptop or you were a SaaS provider serving lots of individual users -- somehow DCR in the initial spec iteration seemed like a good idea (spoiler: it wasn't) and fortunately, the latest revision has somewhat addressed that. XAA/ID-JAG & CIMD should continue to round-out some client management and auth solutions for the enterprise.
Gateways are another area that needs to be addressed in the spec. There isn't a formal definition of one in the spec, and yet, there are lots of "gateways" out there. What a gateway is and what it should do is an open question and it means different things to different people depending on who you ask. For example, who does token exchange: the MCP server or the MCP gateway? Both are valid answers right now depending on the implementation or your opinion of which is best.
More spec iterations should be coming this year. I'm still pretty optimistic about MCP as a whole, as it remains a good way to standardize tool calls across agents and some of the other entities that it provides like resources and prompts are genuinely useful to add more determinism to an agent. Interceptors and skills will be good, too.
If you're interested in helping to evolve the spec further, the MCP Contributors Discord is active. There are lots of IGs/WGs that solicit feedback and you can participate in meetings with your feedback.
Hey thanks for the note about the discord!
I have also been finding the MCP auth story to be really lacking was excited to see OAuth 2 support until I tried to get it to work at work and realized our idp implementation didn't support 2.1, and went into the spec and started wondering if anyone had a good experience yet. Luckily most of our environment can settle on a OAuth token env var standard until that's all in order.
A lot of how well it works or won't work depends on your clients, as not all clients have support for things like RFC 9728 (Protected Resource Metadata). Assuming your client has good support for most of the OAuth 2.0 standards that MCP uses (you don't need DCR as you can statically register clients, assuming that is viable for your environment), then it is possible now with most IDPs to get an OAuth 2.0 auth code flow working just fine. You would then do a token exchange to the upstream to ensure to get the appropriate new audience and rescope/downscope as necessary. Gateways can also help here a lot as instead of baking in all of the auth concerns into your MCP Servers and upstreams, you delegate that to the MCP Gateway. Again, gateway here means different things to different vendors, but Kong, for example, has the ability to act as an MCP proxy (gateway), expose tools based on consumer role or group, apply OAuth 2.0 to it and do an upstream token exchange, while also acting as a regular API gateway that can protect an endpoint with OAuth 2.0/OIDC.
Excellent. Tiny affordances like this should be more commonplace, IMO.
I think this probably also helps when truly clueless users drop the link into an agent directly, because then the agent will relay the message to the user.
> I did something a little bit hacky: if the request is for GET /mcp and the Accept header includes text/html and NOT application/json or text/event-stream, I return a HTML page
Is this not the intended use of request headers?
Technically not really, it's supposed to be alternative versions of the same content.
I think so, but if you wanted to be pedantic you could make the argument that the same resource is not being served with the different Accept header. A welcome page is not the same thing as the JSON returned from the MCP spec.
Maybe it should return a 303 See Other response
A redirect actually does make sense but then there will be some percentage of users that end up copying the redirect URL and try to use that as the MCP URL
I never quite understand why /mcp endpoint is needed.
You can still keep using Rest API with swagger docs and tell the AI to read the swagger docs. It's the same thing. The entire Rest API specification is a lot more flexible than the JSON RPC format /mcp uses.
MCP is fine for exposing tools that don't have APIs, but wrapping an API is an MCP is fine if you like consistency.
Personally, for just tools, I wire up tool calls with context captured via system prompt.
> Despite the fact that MCP is an utterly terrible attempt at a "specification",
Can I just say that anybody involved with MCP's launch should be ashamed of what they put out there. I understand tool calling. I understand specs. I read MCP's "spec" and I used useless word salad that alternates between baby's first wire format and pie-in-the-sky marketing speak. Several of the navigation links I encountered were broken.
Poorly thought out, poorly communicated, but it's the only thing out there that 1) meets the need, and 2) published by people with a huge amount of reach. Right place, right time, shit effort. So it gets adoption. Like the history of PCs, of the internet, or everything, I guess. Worse truly is better.
But why would you show the MCP server URL as a clickable link in the first place, if it's not meant to be clicked? Put it in a monospaced box with a "copy to clipboard" button, it's not the fault of the user for not "thinking ahead" when they click a clickable link that wasn't actually meant to be clicked.
I call this "dev think". Blames user for not thinking ahead when they themselves haven't thought ahead about how users use things. It's everywhere these days, as an end user of things myself, it's maddening.
Yep, unfortunately, not many developers think about good UX.
All kinds of tools make it really difficult to not make a URL clickable and even if it wasn't clickable they might still put it in address bar...
Well, they might also just not read the page the OP made, but a change doesn't have to be perfect to be an improvement.
> I wish the spec had some capability to mitigate for this
I like their solution. It feels like because remote MCP servers are built on HTTP that they are actually using the spec as intended. Serving html when asked for it.
"The (annoying) solution is to package our server up into a connector/plugin and release it for each and every LLM client out there."
Isn't this literally against the entire point of MCP
I hate the MCP rules so much. That I made cookie based auth work by just passing stuff through mimicking the oauth flow.