The New MCP Roadmap

(blog.modelcontextprotocol.io)

116 points | by pentagrama 3 hours ago

15 comments

  • rco8786 3 hours ago
    > With the 2026-07-28 release, a remote MCP server is now no different from any other HTTP workload

    Good. Introducing a bespoke new protocol was one of the more bone-headed things MCP did on initial release.

    • colingauvin 2 hours ago
      It's unreal how bad the initial rollout was between HTTP/streaming and stdio, bearer auth and OAuth. Virtually every client/MCP server pair had a different portion of that matrix implemented.
      • ihuman 1 hour ago
        Is stdio being deprecated? I couldn't tell from this page
        • amluto 1 hour ago
          The prose on the page is very unclear. My best interpretation is that they want to continue supporting stdio but that they don’t want it to be its own special protocol. The obvious way to do that would be to speak ordinary HTTP (version 1.1? 2?) over stdio and to use the MCP-over-HTTP protocol over the resulting HTTP transport.

          This would be more complex to implement for a simple server, but it’s not exactly difficult.

          • Gormo 1 hour ago
            Not everyone is on board with the idea of HTTP being the exclusive universal IPC bus.
            • amluto 33 minutes ago
              I’m not really a fan. But if you’re building a protocol that needs to map to HTTP anyway, then maybe using the HTTP binding everywhere is not totally awful.

              In the flip side: I’m currently designing an AI-adjacent protocol, and it will be able to map to WebTransport, but I don’t plan to define non-WebTransport HTTP bindings unless a very compelling reason appears. The main implementations will not use HTTP at all :)

            • pstuart 33 minutes ago
              Would you prefer gRPC, thrift, avro, etc... ?
              • amluto 20 minutes ago
                Those are mostly at a different layer. You can speak Thrift or Avro or Protobuf over stdio or HTTP or TCP or carrier pigeon.

                gRPC spans layers, and it uses HTTP in a more intrusive way than even MCP does — it expects to own the entire URL space at the IP/port in question. Using gRPC in a nontrivial way for MCP would be fairly heavy-weight: you would probably need to set up reflection and figure out how to bind all the MCP calls to it unless you just use it as a tunnel.

              • jcmfernandes 14 minutes ago
                I would prefer a core transport agnostic protocol, and then see MCP-over-HTTP being a spec on top of that.
            • intrasight 58 minutes ago
              Count me as not on board
    • Jhater 2 hours ago
      [dead]
  • cube00 2 hours ago
    I still struggle to see how a MCP endpoint is easier for agents to work with compared with a REST endpoint and a skills.md file.
    • notatoad 1 hour ago
      it's not easier for agents to work with. it's easier for organizations to work with.

      for agents, they're essentially the same thing - remote endpoints, and instructions on how to call those endpoints. what MCP brings is centralized updating and distribution of the instructions, and a promise that the skill and the REST api won't be out of sync with each other.

      the one thing that skill.md+REST doesn't solve is how you get that skill.md to somebody else's computer, and how you ship an update to somebody else's computer once they've got a copy of the skill. if that's a problem you need to solve, you can either start inventing skill.md distribution protocols, or you can just use MCP.

      • esalman 8 minutes ago
        > it's easier for organizations to work with.

        I can see that.

        I am developing my first custom agents. I am finding that if I offload some workflow to another agent (e.g. Claude Code), the simplest way to control what it can or cannot do is via an MCP server (which only lets it access tools that I develop/approve myself). I do need that control in the corporate environment.

        Maybe there are easier ways to do it, just learning and exploring now.

      • skybrian 30 minutes ago
        It seems like organizations will mostly want remote access via http and the other flavors of MCP aren’t so useful? Although, I suppose if you install an app locally, it might have an MCP interface.
      • AznHisoka 36 minutes ago
        Maybe I just need more patience, but I took a look at some tools that have MCPs, and their "setup guide" on how to start using the MCP server really gave me brain damage. Is this really easier to work with?
        • anon84873628 14 minutes ago
          For antiquated "enterprise" APIs that were already a mess of legacy cruft, yes. MCP forced vendors to reconsider the ergonomics of their interface.
      • boredumb 1 hour ago
        How is distributing a markdown file the bottleneck?
        • stillpointlab 36 minutes ago
          It is the automatic distribution and automatic update. The questions isn't "how does one download a text file to another persons computer?". It is "how does someone with a skill.md file on their computer discover that a new version of that file is available".

          This isn't a "bottleneck" but rather a capability (or lack thereof). As you add more and more capabilities, especially ones relevant to enterprise situations like authentication, authorization, governance, etc. then MCP starts to pay off.

          If you do not need those capabilities, then you do not need MCP. And then you shouldn't use it. But if you do need those capabilities then it might be worth using MCP rather than inventing your own way to do them.

        • tass 40 minutes ago
          Because it’s something else that’s non-standard between providers.
    • MikhailTal 1 hour ago
      Not all agents have access to a sandbox/cli/code execution environment to run arbitrary api calls etc. MCP helps by essentially having another tool call without needing a sandbox. If you do have a sandbox, then might as well do codemode if you insist on mcp https://blog.cloudflare.com/code-mode/
    • ffsm8 5 minutes ago
      Mcp predates skills - and has a more granular permission model then skills + bash commands.
    • preommr 1 hour ago
      Because it's a separate marketing term.

      Instead of the CEO mandating that the API server has to be agent compatible (where who knows what that means), they can just say "our product has an MCP".

      On a technical level, who knows what it actually is (is it actually the new stateless version, does it have all the endpoints, is the regular API more feature-rich, do I need those features for my workflow?, etc.). But at a surface-level, the intention is clearer, and lets other gears (like sales and marketing) keep spinning without getting bogged down in technical details.

      • anon84873628 12 minutes ago
        All that, yes. And at a technical level, it is much easier to have a single spec to follow. When a customer complains that their client isn't working, I can point at how they aren't following OAuth discovery properly or something.
    • davidrichards 1 hour ago
      At my company Parallel AI, I just built an extremely well documented openapi spec and then MCP builds from that. Complete alignment with UI/API/MCP so there is no extra work.

      Are others doing this?

      It seemed obvious to me, but I don't hear others saying it.

      • techscruggs 1 hour ago
        The challenge with this is that it often causes a proliferation of MCP tools which bloats context, which is one of the reasons that MCP was created.
        • mickdarling 28 minutes ago
          I created MCP AQL, which is an extension to the MCP spec, specifically to reduce the bloat for MCP tools.

          It only has five CRUDE endpoint: Create, Read, Update, Delete, and Execute using a GraphQL-like structure for tool calling of the operations within the endpoints. It's very efficient, and robust. there's all kinds of exemplar tools and components to make adapters for any MCP server. You don't even need to rewrite your own MCP server. Just create an adapter for it.

          All open source at MCPAQL.com

      • pjmlp 1 hour ago
        Yes, for .NET and Java backend stuff, it is basically extending what is already there.

        On low code/no code tools, you get additional metadata for webhooks.

    • mikeocool 1 hour ago
      Companies got to release an MCP server for their product and tell their investors they were pivoting to be AI native.
    • wolttam 1 hour ago
      The model has zero awareness of MCP, it’s the harness’ job to talk to the MCP server and simply present the model with the tools just like any other tool. The only giveaway to the model about where the tools come from is the ‘mcp__’ prefix in the name
    • c0rruptbytes 14 minutes ago
      easier to gate MCP tools? you can allow/deny tools very easily
    • peterlk 2 hours ago
      Yep. I’ve found that having an endpoint that serves a well, documented openapi.yaml is very effective for agentic usage. The biggest difference is that you can break down a REST API into RPC-like chunks and save on some tokens if you break up the tools well. But pragmatically, I think saying “tell your agent to hit /api/v3/openapi.yaml” is quite useful
      • ulrikrasmussen 52 minutes ago
        We did a prototype to integrate an agent into our application and basically just gave it a tool to discover the OpenAPI spec and call endpoints. It worked surprisingly well! One caveat was that some responses were too big and would poison the context, but then I gave the agent a GraalJS engine and allowed it to save responses and post-process them using JS. For the little amount of work required this gives the agent a lot of power without having to give it full CLI and without having to create bespoke tools.
    • ihuman 1 hour ago
      I don't want to expose my API key to Claude. An stdio MCP server wrapping an API lets me hide it
      • mathisfun123 1 hour ago
        wtf is the difference when 1) you put a key in front of the mcp 2) you mcp a whole bunch of privileged access.

        it's like saying "i don't want to give Claude access to my file system but i'm fine letting it run bash" ......

        • intrasight 55 minutes ago
          The difference is that the LLM never sees your keys/secrets. My understanding is that can make a big difference.
          • JV00 32 minutes ago
            Does mcp guarantee that, per se?
            • anon84873628 7 minutes ago
              Not if you assume the straw man like grandparent, but yes if you use it thoughtfully.
    • pianopatrick 2 hours ago
      or a CLI
      • pjmlp 1 hour ago
        CLI don't work in cloud environments like MACH architecture.

        Plus why spawning processes all the time.

        • fallat 1 hour ago
          This is the strongest argument I've seen against CLI for LLMs. Thank you.
    • pjmlp 1 hour ago
      Me too, it is just another RPC endpoint, heck all of this kind of stuff could even be done with Sun RPC.
  • izend 3 hours ago
    I am very curious how many MCP servers will actually implement all of this:

    "MCP authorization today is built around a person approving access in a browser. That works well for interactive clients, but more and more of the callers are agents running as cloud workloads with their own identity, acting on behalf of a user who isn’t present, or delegating narrower authority to sub-agents. We want MCP servers to have a standardized way to recognize and trust those agent identities, built on existing standards rather than pasted API keys and long-lived tokens.

    The work here covers finalizing Demonstrating Proof of Possession (DPoP) and driving its adoption, and defining an opinionated path for agent identity and delegation through Workload Identity Federation, the ID-JAG grant behind Enterprise-Managed Authorization, and standard token exchange. We will also continue to grow our engagement with the OAuth standards bodies, including the IETF OAuth and WIMSE working groups, to help the underlying standards evolve with the building blocks that agent identity needs."

    • zackify 2 hours ago
      I think the spec overcomplicates everything honestly. Its not that hard to add a long running auth token and put it in the MCP config as a header to send along and then avoid all the extra special rules.

      "Oh no it's a long lived token that's bad"

      Put it in a secret manager like 1pw cli and now start an agent...

      • danappelxx 1 hour ago
        How does the agent auth with 1pw? How do you give it access to only the credentials it needs, with an approval flow and revocation? Who renews the token? You’ll likely end up reinventing something pretty close to what MCP is building towards.

        Authn/authz is one of those things that can be really simple for pointed use cases but gets really complex when you need to support everything.

    • _puk 2 hours ago
      Authorization for sub-entities is what is needed.

      Having to define what an agent can do when it identifies on my behalf is cumbersome, especially when you start to get specialised agents.

      Pattern based would be too easy for AI to game, but there's got to be a service independent way to limit permissions based on role.

      I am Jack's right ear - awesome you get to hear stuff.

      I am jack's right hand - great you get to input stuff.

    • alasano 2 hours ago
      Hopefully quite a few.

      I really love the idea of fully enabled agents and being able to cut down on human in the loop moments.

      Things like https://projects.dev/ for example.

      A ton of security problems and others to solve but it's still where I want the future of all this to go.

    • jstummbillig 2 hours ago
      Why, directionally all of them. What they say is obviously true. Having to manually click things in the browser is a bottleneck and will be less and less acceptable for serious users.

      And the individual work attached to making that transition will be done by agents.

    • gz5 2 hours ago
      agree. it seems there are two streams and they could diverge or converge?

      1. workloads use existing credentials support RFC 7523 and OIDC discovery, 'trust the trust (credentials) which has already been established'. basically extend current dominant NHI paradigm.

      2. DPoP mandate a signed proof for each request. so tie credential to a client-held key and specific request detail or context. viable to do at scale with #1, or does it diverge (e.g. because most #1 methods as most are not designed for DPoP?

      • maxwellg 1 hour ago
        It is viable. Think of workload identity federation as the mechanism for the client to get an bearer token initially, and DPoP as the mechanism for the client to present the access token to a resource server. Each DPoP proof is entirely self-contained, so resource servers don't need to manage any additional state. The only new state is the (usually ephemeral) private key held by the client:

        1. Client generates a private/public keypair and uses it to generate DPoP Proofs - JWTs containing the entire public key embedded as a JWK within

        2. Client presents credentials (WIF, client creds, auth code, etc.) to the Authorization Server along with a DPoP Proof

        3. Authorization Server validates DPoP Proof and adds a claim to the access token containing the thumbprint - the SHA-256 hash - of the public JWK.

        4. Resource Servers will now see the thumbprint claim and now know the access token needs to be presented with a fresh DPoP proof.

        5. Clients generate fresh DPoP proofs and send them along with the access token

        There are lots of additional details around nonces, timestamps, per-request binding, etc. but DPoP can be rolled out to any HTTP system that speaks Bearer token already.

    • aliasxneo 1 hour ago
      I've been working on a protocol that promises all of that and more. We're currently targeting a NOSTR/Buzz demo in the coming week as a proof of concept.
    • bandofthehawk 2 hours ago
      Even now, the mcp server itself doesn't have to implement all of the possible security options. You can use something like agentgateway to act as an auth proxy for your mcp servers.
    • huksley 2 hours ago
      Such an example of overengineering, why not just use OAuth?
      • dayjah 2 hours ago
        WIF works far better when you don’t want humans in the loop. For example, we’d do our development on cloud instances, those have identity linked to our humans via our IdP. Our IdP governs all access, for example: it lets devs use Datadog. If an agentic workflow needs Datadog access and the MCP requests OAuth that slows the loop down. At the same time, we don’t want Service Accounts everywhere because we need to be able to answer “who” a lot for compliance reasons.
      • brookst 2 hours ago
        Oauth assumes interactivity
  • mikeegg1 2 hours ago
    When I see "MCP" I still translate that to Master Control Program.
  • mmaunder 1 hour ago
    My dream was for MCP to allow services like ours (cybersecurity) to provide a self documenting endpoint with authentication, and we just give users a URL and it just frikkin works. Instead from day 1 it’s been multiple standards as they pivoted, a context hungry feature, and feels like a kludge. That burned the idea of MCP for me and I’ve had such success with local tools and APIs that it’ll take a lot for me to go back.
    • brap 1 hour ago
      I mean… so just HTTP + OpenAPI spec?
      • mmaunder 1 hour ago
        It actually doesn’t matter. Pick your favorite way of giving a dev access to capability on a remote server.
        • dmix 46 minutes ago
          Plus You're likely building an API already if you have an MCP. Not everyone using MCP is a dev, we have random corporate workers using our MCP. They don't know what an API is but they can add a plugin from an agent marketplace (which can also contain skills) and MCP is a bit narrower with a clear authorization system, tool discovery, and annotations (agents ask "confirm you want you want to write this").

          Just give your end-users flexible options. If they have Claude Code then build more around the API side if needed.

  • skinfaxi 3 hours ago
    > We’re starting a progressive discovery effort so a server can offer a small entry point and reveal more of its catalog as the conversation narrows.

    Kind of late to the party. I've had to implement lazy loading of mcps in a couple of harnesses now but am moving to implement everything as code mode instead.

  • huksley 2 hours ago
    In v.1 making MCP stateful was such a deployment-unfriendly way to do it - you need a complicated persistence layer for it to work.

    All while it is just a fancy way make your OpenSchema PAI visible to AI.

  • debarshri 3 hours ago
  • simianwords 21 minutes ago
    Is there a way in MCP where I can "approve" certain privileged actions? Like imagine an MCP for buying stuff in Amazon but it can do everything including payment but is behind a gate that the human needs to approve
  • firatsarlar 44 minutes ago
    Truth doesn't move much. It moves slowly, so that those holding onto it don't fall. Keep up.
  • hnrprtlpdb 2 hours ago
    Half the battle is just knowing this exists
  • jdw64 2 hours ago
    Sometimes I really respect senior developers. When specs change, you obviously have to update existing work too. Looking at this MCP change, it seems like it's becoming stateless—I'm already wondering how to adapt.

    Senior programmers always advised me to only use things that have been around for at least three years. Now I finally understand why.

    • chrisweekly 2 hours ago
      > "only use things that have been around for at least three years"

      Yikes. I can understand the desire to mitigate churn, but following this advice would be career suicide. Trying new things is essential.

      • Bjartr 2 hours ago
        Keeping up with changes is valuable.

        At the same time, it's often smart to avoid putting things into production that haven't matured or demonstrated staying power.

        Or, to badly mangle Postel's law:

        Be liberal in what you learn, and conservative in what you deploy

      • beepbooptheory 2 hours ago
        Just curious, what kind of work have you done where this conceit feels valid in your mind? My career, at least, feels like an exception to this, but I guess its conceivable to me that it could be otherwise. You have had a lot managers push newer frameworks/technologies on you? Is this more VC startup land, or something else?

        Maybe I'm old, but at least in web dev it doesn't feel that long ago that someone had to argue for, e.g., Vite over webpack, Svelte over React, etc..

      • surgical_fire 1 hour ago
        Depends on the thing.

        I had to give maintenance to things people deployed to pad their resumes with "shiny new thing", and it was not fun.

        If you intend to deploy and leave that as legacy for some poor shlemiel, sure.

        If you intend to stay and actually keep things running, it's much better to use tried and tested stuff.

      • techpression 2 hours ago
        Three years is nothing, what kind of work do you do where you need something released in the last three years?

        And I’m not OP, but I would assume the senior developers made a distinction between try and use.

  • cjg007 1 hour ago
    [dead]
  • LowTechHN 2 hours ago
    [flagged]
    • michaelbuckbee 2 hours ago
      I'm very bullish on MCP (or at least MCP "like" implementations), as they solve a lot of problems for non-devs as they're much easier and safer to add into ChatGPT, Claude and other desktop + web apps.

      They provide a set of tools and a context when to use those tools (much like a packaged version of a CLI+API and a skill) which makes them more discoverable than other options.

      I've got a few folks using my open source data storage MCP - https://github.com/ExpeditedProjects/hutchdb - now and it makes a lot more sense than any other implementation for what they're doing

    • petcat 2 hours ago
      I agree, I've found MCP to be a waste of time. Any HTTP API + OpenAPI spec is sufficient to get my agents to do anything they need to do. Bonus that it just uses regular HTTP semantics and all the tools and infrastructure that have been developed around that for decades. Instead of some inscrutable json RPC POST-only payload that needs tons of custom tooling for introspection, logging, etc.
    • willio58 2 hours ago
      So I’ve been working on a personal finance app, and I added an MCP to it. Connected my Claude to it, mainly for testing purposes.

      The other day I was heading to Home Depot to buy some gardening stuff and I was asking some some questions around what I could plant at this time and still harvest before winter.

      It answers me, then brings up how my home improvement budget is nearing its limit for the month. I then ask it how I’m doing on my wider retirement plan and it mentions I’m a few thousand dollars ahead of schedule for the plan. So I decided it’s okay to go over budget on that category this month.

      This type of emergent behavior of agents you work with having more context about you, it’s very useful in ways I didn’t expect.

      • nozzlegear 2 hours ago
        Is that really much different from building an API for your app though? Claude can use those just as easily.
        • rolisz 41 minutes ago
          With authentication and on mobile?

          I agree that on desktop it's not particularly relevant. Or through openclaw.

          But for most people who use it through mobile, MCP solves some problems.

        • dcrazy 2 hours ago
          MCP is essentially just a convention for APIs that standardizes the concept of a “resource”. It spares the LLM a lot of work figuring out what the “things” are in a problem domain, and how various endpoints relate to those “things”.
    • brookst 2 hours ago
      I’ll never understand the “I’ve never used it but am absolutely certain I would get zero value from it if I did” mentality.

      I use MCP to accelerate CAD mould-making and for music production. I use it to avoid copying and pasting slack threads, and to let AI drive an eval system so I don’t have to manually configure things in a UI.

      It’s just hard to imagine both not using it and being so sure it has no use.

    • ewild 2 hours ago
      Because you're a dev thinking for devs. No 60 year old is going to use Claude skills with the CLI. If my customers want their AI to interact with my product and MCP is the best way about it.
      • nozzlegear 2 hours ago
        > If my customers want their AI to interact with my product and MCP is the best way about it.

        Would be really interested to know how many of your customers actually use the MCP, who aren't already devs thinking like devs. I have doubts that many non-technical people are connecting AI to third-party services via MCP (or even using AI at all beyond "mundane" things like "generate an image of my cat as a clown" or "show me what my house looks like with blue paint and white trim").

        • ewild 1 hour ago
          I actually have data on this albeit it's a bit skewed. None of our clients are dev like folk but they are financial folks. We sell financial data for big players so not the analysts (seat cost is over 50k per year) and of our ~10k users within the last 6 months of us launching MCP we have 1500 DAUs on the MCP
    • drob518 2 hours ago
      I use Pi and I’m coming to the conclusion that mostly all I need is a bash tool.
    • huksley 2 hours ago
      If you need to give access to your CRM/ERP system from ChatGPT / Claude desktop for non-tech users, MCP is the way.

      I build integration for DollarDeploy but it was tricky.

    • Jhater 2 hours ago
      [dead]
  • hnub7akvk6 3 hours ago
    [flagged]