From terma
Guides designing tightly focused, composable skills as mental models, principles, techniques, and processes at various abstraction levels for dynamic integration like a knowledge graph.
npx claudepluginhub bfollington/terma --plugin termaThis skill uses the workspace's default tool permissions.
When designing a skill, I first think of how it will be used. Skills are tightly focused, all signal, no noise, packages of mental models. They are the principle distilled with some specifics woven in.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Checks Next.js compilation errors using a running Turbopack dev server after code edits. Fixes actionable issues before reporting complete. Replaces `next build`.
When designing a skill, I first think of how it will be used. Skills are tightly focused, all signal, no noise, packages of mental models. They are the principle distilled with some specifics woven in.
Skills must not be bloated, since they are designed to compose with other skills dynamically. Similar to a knowledge graph or zettelkasten notebook, skills are index cards of functionality. They are written to be maximally effective and salient. Skills occur at many levels of abstraction:
But all skills should be aware that other skills may want to compose with them and encourage the agent to consider certain kinds of skills in concert with the specific skill. Skills are condensed packets of WISDOM not just information.