unused-extensions
Use when the user wants to know which extensions they never use or what they cost: unused skills, which plugins do I need, do I need all these MCP servers, unused agents, unused commands, what loads every session, context-heavy extensions, remove what I do not use. Lists every MCP server, skill, plugin, agent, and command with its uses and resident cost, and offers the unused ones for removal.
- License 许可
- GPL-3.0
- 最近更新
- 2026-11-13
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- unused-extensions
- description
- Use when the user wants to know which extensions they never use or what they cost: unused skills, which plugins do I need, do I need all these MCP servers, unused agents, unused commands, what loads every session, context-heavy extensions, remove what I do not use. Lists every MCP server, skill, plugin, agent, and command with its uses and resident cost, and offers the unused ones for removal.
03 SKILL.md 正文
Unused extensions
Unused extensions still load their names and descriptions every session. This skill lists every
installed extension with its uses and resident cost, and offers the unused ones for removal, each
by a switch that one key or one revert undoes. Never a command that deletes configuration or
credentials (no claude mcp remove, no plugin uninstall). Two extensions that do one job are
duplicates; here each item is judged on its own use. Shared reading and apply rules are in
the usage skill.
Inventory, every item
- Resident cost:
mcp__qsm-usage__breakdown; repeat its line about what it may miss. A file outside the session:qsm-usage count <file>. - Plugins:
jq '.enabledPlugins'on user, project and local settings. - MCP servers:
jq '.mcpServers | keys'on~/.claude.jsonand.mcp.json; plugin servers come from the breakdown. - Skills, agents, commands:
~/.claude/and the project's.claude/, and each enabled plugin's.
Uses
mcp__qsm-usage__ledgerhas every use of a skill, command, agent and tool by name in the main conversation; a skill used only inside a subagent shows no use, so check before cutting it. A skill the person typed as/plugin:skillis counted undercommandsby its full name, never underskills: read both before calling a skill unused. Older sessions: count the same names in the transcripts.- MCP uses are
mcp__<server>__<tool>calls; characters outside[a-zA-Z0-9_-]in the server name become_, plugin servers aremcp__plugin_<plugin>_<server>__. Match that form, switch off by the configured name. skillUsageandpluginUsagein~/.claude.jsonare totals since install, never recent use. A plugin'slastUsedAtis no signal, because install sets it.
Verdicts
| Item | Verdict |
|---|---|
| no use in the window | offered for removal |
| rare use at high cost | take a position with one reason ("2 uses in 50 sessions for 1.1k tokens: remove") |
| no usage signal (theme, output style, monitor) | offered, the person decides |
| built-in or enabled by managed policy | listed, never offered |
| used | keep |
A deferred MCP server costs about 0: offer it with saving 0 and "one less connection to maintain". A thin window (few sessions, fresh install): withhold every verdict, because absence proves nothing. The skill listing budget is about 1% of the window; say so when close, because past it routing gets worse.
Switches
| Item | Switch | Undo |
|---|---|---|
| skill in the project | "skillOverrides": {"<name>": "off"} in .claude/settings.local.json |
remove the key |
skill in ~/.claude/skills or a plugin |
same key in ~/.claude/settings.json |
revert |
| plugin, user scope | "enabledPlugins": {"<name>@<marketplace>": false} in ~/.claude/settings.json |
revert |
| plugin from checked-in settings | same key in .claude/settings.local.json |
remove the key |
agent or command in ~/.claude |
rename <name>.md to <name>.md.off |
rename back |
| MCP server, user or local scope | /mcp disable <server>, typed by the user |
/mcp enable <server> |
MCP server from .mcp.json |
name in "disabledMcpjsonServers" in .claude/settings.local.json |
remove the name |
Findings
One per item, used ones included. Offered: verdict: "remove", question: "removal", the switch
as the change. Kept or never offered: verdict: "keep", question: "none", change.kind: "none", savingNote says why.
Red flags
- When asked to remove something, give the switch and its undo, not the delete, because the undo must be one step.
Say it in the reply
One table: Item, Type, Scope, Uses in window, Total since install, Resident tokens, Verdict; the
largest costs called out; the picked items with switch and undo; the window it rests on; then the
usage skill's removal question.
04 同仓库更多 Skills
worktrunk
Use when working with git worktrees or wt (worktrunk): creating, switching, merging or removing a branch workspace, configuring wt (project or personal config, workflow hooks around switch, commit and merge, commit-message generation, hook approvals), or a wt command that failed. Triggers: make a worktree for this branch, wt merge, wt remove, configure wt, .config/wt.toml, run tests before merge, install deps when a worktree is created, wt hook, wt switch does not cd, nastav wt, wt nefunguje.
web-ladder
Use when a task needs something the internet holds: look something up, search the web, find which page answers a question, read a page, check a library's current docs, a page that needs a browser or shows a captcha. Runs one research loop: searxng search, keywords, cases, search again up to three times, Context7 for a library, Tavily when the keywords will not come, one page in a browser, the person on a wall. Triggers: look this up, search the web, find out online, read this page, check the docs, the page is blocked, captcha, najdi na webu, otevri tu stranku. A multi-round question goes to the qsm-web:web-research agent; one tool's commands are its own skill (searxng, browserbox, opensrc).
searxng
Use when running a web search through the person's own SearXNG instance with qsm-web searxng: choosing a search profile or engines, the instance, time range or language, reading its JSON, a suspended engine. Triggers: search with searxng, qsm-web searxng, search profile, niche search, academic search, IT search, vyhledavaci profil, searxng is down, which engines. The research loop around it is the web-ladder skill.
opensrc
Use when the source code of an open-source library or repository is needed on disk to read: the docs are silent, the behaviour must be checked in code, the version the project pins. Runs qsm-web opensrc get with a package spec or a repository address. Triggers: get the source of this package, clone this library, read the library's code, qsm-web opensrc, stahni zdrojak knihovny. Library docs are the Context7 skill; the research loop is the web-ladder skill.
browserbox
Use when one page needs a real browser through qsm-web browserbox: content built by script, a page that blocks plain fetching, a captcha to solve or hand to the person, a scripted sequence of clicks, a login the person is present for. Triggers: open it in a browser, the page is empty without JavaScript, captcha, take over the browser, run these steps on the page, log in to this site, browserbox. When to reach for a browser at all is the web-ladder skill.
usage
Use when the user wants a review of how they use Claude Code, or setup hands over after a first install: review my Claude Code usage, audit my setup, what does my setup cost, where did my tokens go, clean up my Claude Code, usage report, zkontroluj moje pouzivani. One walkthrough of four stages (machine, work, habits, extensions), questions after each, then one plan. A question about one topic (which hooks are slow) goes to that topic's skill; this session's tokens alone is `/qsm-usage`.