Anthropic: "Equipping agents for the real world with Agent Skills"
Why packaging procedural know-how as loadable skills beats stuffing it all into the prompt — and how progressive disclosure keeps context cheap.
Skills bundle instructions, scripts, and references that a model loads only when a task needs them, so an agent can carry deep capability without paying for it on every turn. The piece lays out progressive disclosure and where skills sit next to prompts and tools. Good grounding for building real agent capability.
Read the source