Skip to main content

Extensible Skills

The customization layer. SpecWeave builds on Claude Code's native skill system with a clean separation:

  • Core instructions live in SKILL.md, stable and versioned
  • Your project-specific rules live in skill-memories/*.md

Skills self-load their customizations through plain LLM instructions embedded in SKILL.md — a cross-platform approach that tells the skill to read your preferences before executing. No shell execution is required. A typical instruction looks like:

Skill Memories: If .specweave/skill-memories/{skill-name}.md exists, read and apply its learnings.

This replaces the previous shell-based dynamic context injection (DCI) approach for skill memories. DCI (Claude Code's !`command` syntax) is still used for other purposes like loading skill context or PR diffs, but skill memories now use instruction-based loading for better cross-platform compatibility.

A skill that explicitly reads a project preference file can reuse those instructions. That depends on the skill and harness loading the file; this pattern does not guarantee automatic learning or application in every future session.


In This Section​

Extensible Skills Specification​

The formal specification — extensibility category definitions, dynamic context injection specification, detection algorithm, and conformance requirements.

Implementation Guide (historical reference; not published)​

The separate how-to document for the earlier skill-memories architecture is unavailable. This section does not supply its setup instructions or FAQ.

Customization Without Modification​

Overview page with a quick reference table and a link to the published specification.

Claude Skills Deep Dive (historical reference; not published)​

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Self-Improving Skills (Reflect; historical reference, not published)​

Earlier auto-learning design. It is not a current SpecWeave 2.1 lifecycle guarantee; see the current skills reference.

Development Guidelines​

Design principles for skill authoring — how to design skills that users can extend without modification.


See Also​