I still think of MCP as "Master Control Program".
Stateless MCP has recaptured my interest
191–200 of 246 posts
Re: Stateless MCP has recaptured my interest
#192Earlier quoted context omitted.
They didn't reinvent it, it's literally using JSONRPC. The thing they "invented" was coming up with a loose standard for how to implement your RPC endpoints in a way that the AI was pre-trained to interact with. The actual implementation may have been very janky but the idea itself is sound: If you want to interact with arbitrary services then providing one single protocol (ideally not JSONRPC but hey) with a standar…
MCP is invisible to the LLM, or should be, under normal use
Not sure I understand what you mean. MCP servers and methods available do have to be visible to the model in some way so that it can spit out tool calls instructions to be executed by the orchestration layer around it.
But you probably meant something else?
Re: Stateless MCP has recaptured my interest
#193Earlier quoted context omitted.
MCP is invisible to the LLM, or should be, under normal use
Not sure I understand what you mean. MCP servers and methods available do have to be visible to the model in some way so that it can spit out tool calls instructions to be executed by the orchestration layer around it. But you probably meant something else?
MCP is the protocol by which the harness can connect with additional tools. The harness does the work of discovering the tools, the model is then fed a description of what tools it has available, with zero awareness of whether those tools are MCP or built into the harness itself (except for by the mcp__ prefix in the tool’s name, which a harness could choose to strip off)
Re: Stateless MCP has recaptured my interest
#194Earlier quoted context omitted.
Skills don’t need to be preloaded, can be invoked by name on demand; and when you list available skills, a one-line description for each is enough.
And all those "names and one-line descriptions" live in an .md file and of course pollute your context. And then the model will randomly decide to load the .md file that is the "skill" randomly at any point. Again polluting the context. There's no magic.
Most people seem to mean "Skills don't pollute the context window [with their entire contents]", they just elide the part in brackets.
There is no magic and people should be more precise when discussing skills.
Re: Stateless MCP has recaptured my interest
#195It is amazing that folks getting paid several hundred thousand dollars a year...reinvented RPC-over-HTTP/JSON. You too, fellow web developers, are smart enough to work at Anthropic. I would love to see a proper engineering post-mortem for how this happened.
It's still too early stage for me. Anthropic needs to mature for a little longer before they're enlightened enough to understand my kafka-based event sourcing model for cross-agent orchestration. Kafka, protobuf, functional domain modelling using F# as a DSL...the whole kit and kaboodle. Only when my AI and its subagents are modelled through a graph of producing agents, consuming agents, and agentic mailboxes, will C…
I can't tell if this is sarcasm or an advertisement for your start up.
Re: Stateless MCP has recaptured my interest
#196Re: Stateless MCP has recaptured my interest
#197In this thread I’m seeing a couple of the typical skill arguments. Skills don’t pollute the context window and, skills are more flexible and composable. In this specific post the SQL example gets around this by being super low-level, inherently composable, and low context consumption itself. I’m not sure this works for everyone. Looking at the Linear MCP for example, they basically just want to expose an API surface…
> Skills don’t pollute the context window and, skills are more flexible and composable.
they do but more importantly model does not give a flying fuck about what you put in there in any consistent way.
ppl have been mislead by ai companies that they've invented some sort of agi that you can 'explain' to using skills. So ppl have come up with also sorts of skills that dont do jack shit.
Re: Stateless MCP has recaptured my interest
#198I think stateless-type MCP was already possible, eg my MCP Clock [ https://github.com/firasd/mcpclock ]: > curl -s -X POST "https://mcpclock.firasd.workers.dev/mcp" -H "Content-Type: application/json" -H "Accept: application/json, text/event-stream" -d '{"jsonrpc":"2.0","id": 1,"method":"tools/call","params":{"name":"clock_get","arguments":{}}}' | grep '^data:' | sed 's/^data: //'| jq {"result": {"content": [{"type":…
I am not sure that coding doesn't lead token usage.
Re: Stateless MCP has recaptured my interest
#199It is amazing that folks getting paid several hundred thousand dollars a year...reinvented RPC-over-HTTP/JSON. You too, fellow web developers, are smart enough to work at Anthropic. I would love to see a proper engineering post-mortem for how this happened.
In fact this aspect of "re-inventing and then congratulating oneself for it" has been a factor in software since the dawn of the subject - it is now with AI/ML, becoming far more endemic across a far broader group of people, who ordinarily would not have done software development properly if it weren't for the immense power that comes from glib application of tooling, minus methodology. I have long observed that tool…
It used to be partially because building tools is fun. Often more fun than building whatever the company actually needs. But now it’s not that fun.
LLMs need to say stuff like “your company sells sparkling water, you should not build and maintain another in-house product board and CRM”, and then stand their ground.
Re: Stateless MCP has recaptured my interest
#200Earlier quoted context omitted.
The "CLI crowd" is also primarily using LLMs on their own computer. Where they have their CLI tools. This doesn't cover the case when you're talking to an LLM from web, or via Slack or Linear, etc. There, you will want MCP so the LLM can use services on your behalf as you. That's portability.
Why can’t the LLM you’re talking to on the web have access to your CLI tools? When you talk to an LLM on the web, the harnesses spin up a fresh environment (I would hope it’s a VM…) so that the LLM can do stuff like run arbitrary Python and Bash scripts to complete the task you asked it for. There’s no reason why you shouldn’t be able to customize this environment to add whatever CLI tools and credentials you need fo…
Could be more expensive to provide a computing environment.