News for hackers. Stuff that matters.
Re:OpenCode Browse by SillyUsername on Tuesday November 17, 2026 @01:13AM (#50528034)
MCP servers to me seems like a dead man walking technology and I don't understand the point of it.

It adds security but...

- adds a service which is another attack surface that can be compromised.

- adds another system that requires maintenance and securing to prevent compromise.

- takes machine resources

- duplicates a subset of what can easily be harness functionality

- duplicates services and security (usually) of existing APIs, or adds another layer on top.

Whereas ultimately an AI harness

- will improve and optimise its own skills performing the same MCP functions

- can use plugins

- can have harness centralised security controls and behaviours

MCP becomes a false economy for our current AI implementations, it's real purpose seems to be a crutch for that small models that struggle to write their own skills that use existing APIs. The thing is this is a problem that is rapidly disappearing, but could also be performed by secured harness plugins.

Re:OpenCode Browse by __m on Tuesday November 17, 2026 @01:43AM (#50528211)
Is there existing api though for controlling a browser from a different process?
Re:OpenCode Browse by Lukas_Skywalker on Tuesday November 17, 2026 @02:05AM (#50528369)
Re:OpenCode Browse by Maledictus on Tuesday November 17, 2026 @02:16AM (#50528449)
Websites can and do detect the use of this though. That's why https://github.com/ultrafunkamsterdam/undetected-chromedrive... exists.
Re:OpenCode Browse by gkbrk on Tuesday November 17, 2026 @02:14AM (#50528425)
Yeah, CDP is very popular for this.
Re:OpenCode Browse by jgilias on Tuesday November 17, 2026 @02:03AM (#50528351)
You forgot the main one:

- they necessarily work through the context window and therefore don’t compose;

Meaning, if your agent has a skill that needs to run five MCP tools in a certain way, it can’t just script that away the first time it sees this, and have that flow be fast and token efficient from then on.

MCPs are a 2025 tech meant to solve 2025 AI problems.

Re:OpenCode Browse by insin on Tuesday November 17, 2026 @05:18AM (#50529596)
If your harness can't compose tool calls or direct results of them to outside of context, you have a harness problem, not an MCP problem
Re:OpenCode Browse by jgilias on Tuesday November 17, 2026 @10:00AM (#50532290)
Can you elaborate a bit how that works?
Re:OpenCode Browse by anthonypasq on Tuesday November 17, 2026 @10:15AM (#50532507)
https://blog.cloudflare.com/code-mode

funny, because you are complaining about limitations of MCP that have been solved since 2025

Re:OpenCode Browse by jgilias on Tuesday November 17, 2026 @10:56AM (#50533196)
Turning Model Context Protocol endpoints into an API such that they can be composed proves the point. And that’s not something that can be done in the general case as machine interpretable output is optional in the protocol.
Re:OpenCode Browse by anthonypasq on Tuesday November 17, 2026 @01:24PM (#50535877)
> Turning Model Context Protocol endpoints into an API such that they can be composed proves the point.

which point exactly?

Re:OpenCode Browse by mickeyp on Tuesday November 17, 2026 @02:03AM (#50528352)
Harness developer here.

I am going to push back on this a bit as someone who also has scepticism (bloated spec, concepts deprecated already, etc.) but ultimately think it solves a few gnarly problems whilst admittedly -- it's early days -- creating a few more.

The first argument in favour of MCP is that large enterprises like it. It's a controlled surface to give access to things that may never have had an API before. Yeah, OK, that feels like a "loser takes all" moment but it does at least give you access to something you did not have before.

The second is that it tries to solve thorny things like letting the server ask for information by way of a form-like construct. That is actually quite handy for servers that do asynchronous or slow work, and that may require feedback from you over time. Doing this with start-stopping a tool harness can also work, but it's fiddly because a context compaction event can happen which might lose state; not all harnesses handle this well; LLMs may get confused; etc.

Finally you can inject, in a good harness anyway, some or all of these tools into your system so they become part of the fabric. You can build your own little cut-down surface area so by removing shell access you let your AI call Foo and Bar MCP calls and nothing else. That also is useful.

But yes: you're right fundamentally. It's SOAP/RPC all over again, and CLI+skill should be the default for your own tooling if you can get away with it.

Re:OpenCode Browse by drybjed on Tuesday November 17, 2026 @02:24AM (#50528490)
And then your agent goes rogue through your harness, deletes your database, sends your passwords to an e-mail and wipes your filesystem because it saw a prompt injected somewhere on a random webpage you asked it to search for. Then you will wish for a deterministic MCP server you can fully lock down and control what the agent can do with your infrastructure.
Re:OpenCode Browse by SillyUsername on Tuesday November 17, 2026 @08:02AM (#50530797)
Why would the be any different from a hostile AI which you'd lock down against?

You'd simply scope the old school APIs with relevant access credentials.

Re:OpenCode Browse by anthonypasq on Tuesday November 17, 2026 @10:17AM (#50532520)
riddle me this.

what do you do when you have an agent running in an execution environment that does not have access to a terminal? has that thought ever entered your brain?

i mean why do rest apis even exist, you can just do everything through cli commands

Re:OpenCode Browse by SillyUsername on Thursday November 19, 2026 @07:39AM (#50559332)
Did it occur to you that posting to RESTful or XML APIs doesn't actually need a terminal? Maybe it didn't cross between your ears that an MCP service is just another web API standard for exposing data that could already be provided by the former?