From style-switcher
Add a named style snippet to a category in the user's library. Creates the category directory if it doesn't exist. Snippets are short tonal fragments — register, terseness, hedging, formatting — meant to be layered together via apply-layers.
How this skill is triggered — by the user, by Claude, or both
Slash command
/style-switcher:add-snippetThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
1. Resolve `$DATA_ROOT` from `config.json`. If config is missing, run `onboard` first.
Resolve $DATA_ROOT from config.json. If config is missing, run onboard first.
Ask:
verbosity, initiative, tone). List existing categories from ls $DATA_ROOT/snippets/ and let the user pick or type a new one. Create the dir if new.ultra-curt, free-spirit) — must be unique within the category. Reject if $DATA_ROOT/snippets/<category>/<name>.md already exists (offer edit-snippet).list-snippets).Write to $DATA_ROOT/snippets/<category>/<name>.md:
---
name: <name>
category: <category>
description: <one-liner>
created: <ISO-8601>
---
<body>
Confirm with the path, and ask whether to apply now (delegates to apply-layers to update <category> to this snippet).
A snippet is a subtle composable fragment — a few sentences that nudge tone in one specific dimension. Think:
Not a persona, not a behavioral override, not a multi-paragraph manifesto. If it reads like a system prompt, it's too heavy.
Second-person ("You..."), present tense, declarative, focused on how to phrase. Two to four sentences max. No meta-commentary.
npx claudepluginhub danielrosehill/claude-code-plugins --plugin style-switcherProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
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.