Claude Skills are awesome, maybe a bigger deal than MCP

#1

Post by weinzierl » Fri, Oct 17, 2025, 5:40 PM UTC

Claude Skills are awesome, maybe a bigger deal than MCP
simonwillison.net

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#2

Post by billconan » Fri, Oct 17, 2025, 5:45 PM UTC

> LLMs know how to call cli-tool --help, which means you don’t have to spend many tokens describing how to use them—the model can figure it out later when it needs to.

I do not understand this. cli-tool --help outputs still occupies tokens right?

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#3

Post by SoMomentary » Fri, Oct 17, 2025, 5:49 PM UTC

billconan wrote: post #2

> LLMs know how to call cli-tool --help, which means you don’t have to spend many tokens describing how to use them—the model can figure it out later when it needs to. I do not understand this. cli-tool --help outputs still occupies tokens right?

Absolutely, but it occupies them later and only when needed. This is what I think they're driving at here.

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#4

Post by billconan » Fri, Oct 17, 2025, 5:59 PM UTC

SoMomentary wrote: post #3

billconan wrote: post #2

> LLMs know how to call cli-tool --help, which means you don’t have to spend many tokens describing how to use them—the model can figure it out later when it needs to. I do not understand this. cli-tool --help outputs still occupies tokens right?

Absolutely, but it occupies them later and only when needed. This is what I think they're driving at here.

but why can't I do the same with mcp? I just create a help() function that returns the help info?

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#5

Post by brazukadev » Fri, Oct 17, 2025, 6:02 PM UTC

billconan wrote: post #4

SoMomentary wrote: post #3

Earlier quoted context omitted.

Absolutely, but it occupies them later and only when needed. This is what I think they're driving at here.

but why can't I do the same with mcp? I just create a help() function that returns the help info?

Most people still don't understand MCP properly and think it's about adding 50 tools to every call. Proper MCP servers and clients implement tools/listChanged

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#6

Post by 8note » Fri, Oct 17, 2025, 6:03 PM UTC

billconan wrote: post #4

SoMomentary wrote: post #3

Earlier quoted context omitted.

Absolutely, but it occupies them later and only when needed. This is what I think they're driving at here.

but why can't I do the same with mcp? I just create a help() function that returns the help info?

you can; i've seen people put mcp access behind another mcp. I'm not sure how much success they got from it though

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#7

Post by CharlesW » Fri, Oct 17, 2025, 6:05 PM UTC

billconan wrote: post #4

SoMomentary wrote: post #3

Earlier quoted context omitted.

Absolutely, but it occupies them later and only when needed. This is what I think they're driving at here.

but why can't I do the same with mcp? I just create a help() function that returns the help info?

That hypothetical might be fine, but MCPs do much more than that and their catalogs can be enormous. Here are some popular MCPs and the amount of context they eat before you've done anything with them:

  • Linear: 23 tools (~12,935 tokens)
  • JetBrains: 20 tools (~12,252 tokens)
  • Playwright: 21 tools (~9,804 tokens)

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#8

Post by jamesclar » Fri, Oct 17, 2025, 6:06 PM UTC

I have been using Claude since few month and couldn't switch to another one regarding performance

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#9

Post by fullstackchris » Fri, Oct 17, 2025, 6:07 PM UTC

I just read this seperately through Google Discover, and I don't quite get amazing newness of it - if anything, it feels to me like of an abstraction of MCP - there is nothing I see here that couldnt be replaced by a series of MCP tools - for example, the author mentions "a current trick" often used is including a markdown file with details / instructions around a task - this can be handled with an mcp server prompt (or even a 'tool' that just returns the desired text) If you've fooled around as much as I have, you realize in the prompt itself you can mention other available tools the LLM can use - defining a workflow, if you will, including tools for actual coding and validation like the author mentions they included in their skill.

Furthermore, with all the hype around MCP servers and simply the amount of servers now existing, do they just immediately come obsolete? its also a bit fuzzy to me just exactly how an LLM will choose an MCP tool over a skill and vice versa...

Re: Claude Skills are awesome, maybe a bigger deal than MCP

#10

Post by frankc » Fri, Oct 17, 2025, 6:09 PM UTC

So far I am in the skeptic camp on this. I don't see it adding a lot of value to my current claude code workflow which already includes specialized agents and a custom mcp to search indexed mkdocs sites that effectively cover the kinds of things I would include in these skills file. Maybe it winds up being a simpler, more organized way to do some of this, but I am not particularly excited right now.

I also think "skills" is a bad name. I guess its a reference to the fact that it can run scripts you provide, but the announcement really seems to be more about the hierarchical docs. It's really more like a selective context loading system than a "skill".