should-i-use
Honest library picks for coding agents in 25-360 tokens. Tells your agent what NOT to install.
我该使用它吗
质量与安全性
发现（3）
- LOWTool 'alternatives' description lacks action verb在 alternatives 中
- LOWTool 'how_do_i' description lacks action verb在 how_do_i 中
- LOWTool 'docs_link' description lacks action verb在 docs_link 中
基于对工具定义和协议合规性的自动分析。
上下文开销
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"should-i-use": {
"command": "npx",
"args": [
"should-i-use-mcp"
]
}
}
}可运行的软件包
0.3.2stdio远程端点
https://mrkeyoor.com/mcpstreamable-http它能做什么
工具清单
工具（6）
🟢pick_library(task, ecosystem)
Given a coding task, recommend 1-3 libraries from the should-i-use index with one-line reasoning and an honest warning about the top pick. Use before installing anything.
输入模式
{
"type": "object",
"properties": {
"task": {
"type": "string",
"description": "What you are trying to do, e.g. \"parse CSV files in node\""
},
"ecosystem": {
"type": "string",
"enum": [
"npm",
"pypi"
],
"description": "Optional registry filter"
}
},
"required": [
"task"
]
}输出模式
{
"type": "object",
"properties": {
"answer": {
"type": "string",
"description": "The clamped text answer (hard limit ~500 tokens)"
}
},
"required": [
"answer"
]
}🟢should_i_use(library)
Honest verdict on a specific library: 4-axis scores with reasons, when to skip it, and maintenance signals (last push, weekly downloads).
输入模式
{
"type": "object",
"properties": {
"library": {
"type": "string",
"description": "Library name as published on npm or PyPI"
}
},
"required": [
"library"
]
}输出模式
{
"type": "object",
"properties": {
"answer": {
"type": "string",
"description": "The clamped text answer (hard limit ~500 tokens)"
}
},
"required": [
"answer"
]
}🟢alternatives(library)
Curated alternatives to a library, each with a one-line "prefer it when" and a verdict if the alternative is also indexed.
输入模式
{
"type": "object",
"properties": {
"library": {
"type": "string",
"description": "Library name as published on npm or PyPI"
}
},
"required": [
"library"
]
}输出模式
{
"type": "object",
"properties": {
"answer": {
"type": "string",
"description": "The clamped text answer (hard limit ~500 tokens)"
}
},
"required": [
"answer"
]
}🟢how_do_i(library, task)
Return the 1-2 best code snippets for a task with a library, plus the gotcha for each. Snippets are correct for the indexed version. Far cheaper in tokens than a docs dump.
输入模式
{
"type": "object",
"properties": {
"library": {
"type": "string",
"description": "Library name as published on npm or PyPI"
},
"task": {
"type": "string",
"description": "What you want to do, e.g. \"retry a failed request\""
}
},
"required": [
"library",
"task"
]
}输出模式
{
"type": "object",
"properties": {
"answer": {
"type": "string",
"description": "The clamped text answer (hard limit ~500 tokens)"
}
},
"required": [
"answer"
]
}🟢audit_dependencies(manifest, ecosystem)
Audit a project's dependencies for unmaintained packages, unstable APIs, and documented reasons to switch. Pass the contents of package.json, requirements.txt, or pyproject.toml. Returns only the dependencies worth a decision, not a report on every line. Run this before working in an unfamiliar codebase.
输入模式
{
"type": "object",
"properties": {
"manifest": {
"type": "string",
"description": "Contents of package.json, requirements.txt, pyproject.toml, or a newline-separated list of package names"
},
"ecosystem": {
"type": "string",
"enum": [
"npm",
"pypi"
],
"description": "Optional hint when the manifest format is ambiguous"
}
},
"required": [
"manifest"
]
}输出模式
{
"type": "object",
"properties": {
"answer": {
"type": "string",
"description": "The clamped text answer (hard limit ~500 tokens)"
}
},
"required": [
"answer"
]
}🟢docs_link(library, topic)
Official docs URL and GitHub repo for a library, one line each. Use when the clamped answers are not enough.
输入模式
{
"type": "object",
"properties": {
"library": {
"type": "string",
"description": "Library name as published on npm or PyPI"
},
"topic": {
"type": "string",
"description": "Optional topic to look up in the docs"
}
},
"required": [
"library"
]
}输出模式
{
"type": "object",
"properties": {
"answer": {
"type": "string",
"description": "The clamped text answer (hard limit ~500 tokens)"
}
},
"required": [
"answer"
]
}社区
证据