What is Agent Skills?
Discrete capabilities that an AI agent can invoke to perform specific tasks within a workflow.
Agent skills are modular, self-contained units of functionality that define what an AI agent can do. Each skill typically includes a description of its purpose, the inputs it expects, the outputs it produces, and any constraints on when it should be used. Skills allow agents to be composed from reusable building blocks rather than monolithic instruction sets, making them easier to test, share, and iterate on. In practice, skills are often declared in structured files like SKILL.md and discovered at runtime by the agent orchestration layer.