←back to thread

237 points jdkee | 5 comments | | HN request time: 1.17s | source
Show context
deadbabe ◴[] No.45948490[source]
You don't need MCP.

You need Claude Skills.

replies(2): >>45948496 #>>45949633 #
CuriouslyC ◴[] No.45949633[source]
Claude Skills are just good documentation wrapped into Anthropic's API in a proprietary way that's designed to foster lock-in.
replies(1): >>45951182 #
1. sunaookami ◴[] No.45951182[source]
How are skills vendor lock-in when they're just Markdown files that any LLM can read? You are not locked to Anthropic at all.
replies(1): >>45952889 #
2. CuriouslyC ◴[] No.45952889[source]
The API runs your skills, it's not client side coordinated. You have to replicate the skill-running behavior locally.
replies(1): >>45953635 #
3. sunaookami ◴[] No.45953635[source]
Claude just reads the SKILL.md frontmatter into initial context and when the instructions match it reads the rest of the SKILL.md. Every LLM can do that.
replies(1): >>45955804 #
4. CuriouslyC ◴[] No.45955804{3}[source]
If it's such a thing that every model already can support it and you don't need to do anything with it, why is it even a feature?
replies(1): >>45956624 #
5. sunaookami ◴[] No.45956624{4}[source]
To be fair other providers have to implement reading a skill directory and only the frontmatter of the SKILL.md and read more of it when needed. But it's a LOT simpler than implementing MCP.