Skip to main content

Extensible Skills Standard

The customization layer. Extensible Skills apply the Open/Closed Principle from SOLID design to AI agent skills:

  • Closed for modification — Core skill logic lives in SKILL.md, stable and tested
  • Open for extension — Your project-specific rules live in skill-memories/*.md

Skills self-load their customizations using Dynamic Context Injection (DCI) — a shell one-liner that reads your preferences before the skill executes. Three-tier cascading lookup ensures project-level overrides take priority over global defaults.

The result: you correct Claude once ("use React Hook Form, not useState for forms"), and that preference is applied automatically in every future session.

In This Section​

Extensible Skills Standard​

The formal specification — E0-E4 tier definitions, DCI specification, frontmatter schema, agent portability matrix, and conformance requirements.

Implementation Guide​

Practical how-to — getting started, architecture, cascading lookup, real-world examples, skill memory format, and FAQ.

The Open/Closed Principle for AI​

Overview page with quick reference table and links to the standard and guide.

Claude Skills Deep Dive​

How skills work under the hood — progressive disclosure architecture, comparison with other AI tool systems, and the evolution from prompts to programs.

Self-Improving Skills (Reflect)​

The Reflect system auto-learns from corrections and saves them to skill memories. Correct once, applied forever.

Development Guidelines​

SOLID principles applied to skill authoring — how to design skills that users can extend without modification.

See Also​