Context window for most frontier models is 1 million tokens. They all start to lose their minds around 300K, if not sooner.

How do I say this in the most gentle way I can...?

Your brain is broke. No, seriously. If you can help it, try not to argue with anyone about anything, ever. You have a demonstrated inability to think clearly.

Many believe, including myself, that the market is currently propped by a massive AI bubble. Nearly a US $1 trillion is being spent this year, and more is planned for next year. All of this is for a "build up". There is no pay out. The major AI companies are taking in massive losses in the hopes that they will eventually be able to cash out.

The math is not looking good to me. The effect will be like the dotcom bubble. But much much bigger. Because the numbers are so much bigger.

I have a skill that describes my database schema with hundreds of tables, stored procedures, triggers at a high level. It tells the LLM to use a specific CLI to connect to the developer database using an account with read-only permissions. This skill has auto invocation turned off and uses zero tokens when not used.

MCP apps do not have a manual invocation option. If you enable them, they are always on, eating away tokens, even if you are not using them at all. And for my simple example above, would be a lot more work to setup. And once setup would eat way more tokens when compared to a small skill file that describes a CLI tool for querying the SQL database.

Output of an MCP tool call goes straight into context. Output of a CLI tool call can be filtered by the LLM using another tool, before it enters context.

These are meaningful differences and show that skills get the work done in an efficient manner, vast majority of the time.

Even when it is disabled? The point is to have the skill or mcp app to not use ANY tokens until I call it by name. I can do that with a skill. Can I do that with an mcp app?

Last time I checked, and it may be different now, Claude Code used tool search only if context usage was greater than 10%.

Also, even with tool search, MCP uses about 1000 to 3000 tokens. A skill frontmatter on the other hand can be around 50.

And an mcp tool call result gets dumped into context in its entirety. A cli too can be filtered with grep/sed etc.

Skills are just a lot more efficient for most use cases.

disable-model-invocation: true

Most MCP servers still bloat up my context even when I am not using them at all. Why can't they be configured to allow me to keep them out unless I really need to use them?

Are you claiming that the frontier labs like OpenAI and Anthropic are actually making a profit contrary to all the claims?

Some of this may be perception rather than reality. I have never felt that ChatGPT was being smug or rude to me.

LLMs are great for this sort of thing that used to be a massive pain in the rear.

I am guessing this is not targeting those of us on the heavily subsidized $200/mo plans. Sure, these plans may be temporary, but none of us really know how temporary they are. Until then, 1/3rd of the published API pricing is not very appealing.

For a SaaS business, running servers isn't free. But compared to the cost of running GPUs for inference that you are selling, it almost is. The company I work for is a SaaS company. We have a single production server. A couple of QA servers. All hosted on Hetzner. Monthly cost for servers is less than $400. This generates a few million dollars a year in revenue.

If we were in the business of selling inference, our cost of providing the service, for the same amount of revenue would significantly higher.

Even large businesses like Microsoft, Meta, Google have operated with similar margins. Cost of running servers, compared to revenue was very low. But inference changed that, in a dramatic way.