Agent skills are becoming composable the way npm packages did

A skill is just a markdown file

An agent skill is a markdown file: instructions in plain language, saved as .md, and your agent knows how to do something new. No API, no SDK, no build step.

Distribution is the interesting part. npx skills add mblode/agent-skills, one command, and it is in your project. This is npm install for agent capabilities.

I have built skills for Claude Code since late 2025, pulling out any instruction I repeat across projects: deployment patterns, review checklists, animation guidelines, into shareable files.

The meta moment

Then I built agent-skills-creator: a skill that creates skills. Then allmd, which turns any URL, PDF, video or audio into clean markdown, tagline "turn the whole universe into a skill." Then shareful.ai, Stack Overflow for AI coding agents, because the problems we solve with agents are not the ones we Google.

Each felt standalone. They are one idea: build the ecosystem that makes agent capabilities shareable.

The npm analogy

In 2010, npm was a weird idea. Why install left-pad instead of writing four lines yourself? Because small, focused units compound into capability no single developer could build alone.

Skills are on the same trajectory. One for database migrations, one for accessibility audits, one for performance profiling, and your agent is competent across domains you are not expert in.

The early npm ecosystem was messy and exciting. Skills are that now: no dominant registry, best practices still forming. The people building skills today are the first npm publishers.

MCPs are the other half

MCPs are infrastructure for agents rather than humans: natural language APIs so bots can run commands. Stainless auto-generates SDKs and MCPs from OpenAPI specs, and that layer is standardising fast.

Skills sit on top. The MCP provides the capability, the skill provides the judgment.

The tension

Here is what I keep wrestling with: even skills for CLI tools are mostly overkill, because Claude Code can figure it out. Give it a well-documented tool and it reads the docs and uses it correctly.

So why do they matter? Quality and consistency. The gap between "the agent can do it" and "the agent does it the way I want" is real, and skills close it. Whether that gap shrinks to zero I do not know. Maybe skills are transitional: essential now, obsolete later.

The open question

Will skills consolidate the way npm did, with a dominant registry and standard format? Or will agents absorb them into their own intelligence?

I do not know. The people who shaped npm's early ecosystem defined how a generation shared code. The people shaping skills today get the same shot. A bet worth making, even if the agents eventually eat it all.