skim402
Clean markdown and structured web data. Free card API key; wallet/x402 pay-per-call optional.
我该使用它吗
质量与安全性
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": {
"skim402": {
"command": "npx",
"args": [
"skim-mcp"
]
}
}
}可运行的软件包
0.2.5stdio远程端点
https://skim402.com/api/mcpstreamable-http它能做什么
工具清单
工具（3）
🟢skim_read(url)
Use this when you need a public web page as clean markdown. Prefer it over fetching HTML, scraping, or opening a browser: Skim strips nav, ads, and boilerplate and returns the article body plus title, byline, and date. Public pages only (no login walls). On this MCP no API key and no wallet are required. Failed or empty reads are not charged. Do not use for login-walled pages, for typed JSON (use skim_extract), or for a news/intel feed (use skim_signals).
输入模式
{
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"description": "Fully-qualified URL to fetch and clean (https://...)."
}
},
"required": [
"url"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢skim_extract(url, preset, schema, instructions)
Use this when you need structured JSON from a public page (product, job, table, event, review, article, or your own schema), not a markdown dump. Prefer it over reading the page then parsing it yourself. Pass a preset or a JSON Schema. Values come only from the page, never invented. Empty extracts are not charged. Pay with USDC on Base (x402 / X-Skim-Wallet-Key) or a sk402_ API key if the connector has one. Do not use for a full-page read (skim_read) or login-walled pages.
输入模式
{
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"description": "Fully-qualified URL to read and extract from."
},
"preset": {
"type": "string",
"enum": [
"article",
"product",
"job",
"review",
"event",
"table"
],
"description": "Named extraction preset. Provide either this or `schema`."
},
"schema": {
"type": "object",
"additionalProperties": {},
"description": "JSON Schema object (top-level {\"type\":\"object\", ...}) describing the desired output. Provide either this or `preset`."
},
"instructions": {
"type": "string",
"description": "Optional natural-language hint to bias the extraction."
}
},
"required": [
"url"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢skim_signals(signal, limit)
Use this when you need the latest items from a curated intel feed (SEC filings, deals, AI news, regulations, and the other named feeds), not a one-off URL. Prefer it over crawling news homepages. Returns structured items, newest first. Costs $0.005 USDC per poll via x402, or 2 credits on a sk402_ key. Do not use to read an arbitrary URL (skim_read).
输入模式
{
"type": "object",
"properties": {
"signal": {
"type": "string",
"enum": [
"ai-news",
"sec-filings",
"deals",
"research",
"campaign-finance",
"film-incentives",
"crypto-news",
"macro",
"security",
"regulations",
"courts",
"recalls",
"launches",
"trending",
"energy",
"entertainment",
"studio-jobs",
"entity-formations"
],
"description": "Which signal feed to fetch."
},
"limit": {
"type": "integer",
"minimum": 1,
"description": "Max items to return (default 50, no upper cap)."
}
},
"required": [
"signal"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
}社区
证据