Skip to main content
Glama
SecretRichGarden

Metaso Search MCP Server

by SecretRichGarden
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Remote

Metaso Search MCP Server | 秘塔搜索 MCP 服务器

MCP server for Metaso Search API, providing search and web reader tools over stdio.

基于秘塔搜索 API 的 MCP 服务器，通过 stdio 提供搜索与网页读取能力。

Features | 功能

  • metaso_search: multi-scope search (webpage, document, paper, image, video, podcast)

  • metaso_reader: read webpage content in markdown or json

  • npm-ready package with CI/CD publish workflow

  • npx config generator for Cherry Studio / Cursor / Claude / generic MCP clients

Related MCP server: bocha-mcp

Install | 安装

From npm (recommended) | npm 安装（推荐）

npm install -g metaso-search-mcp

From source | 源码安装

git clone https://github.com/SecretRichGarden/metasota-API-MCP.git
cd metasota-API-MCP
npm install
npm run build

MCP stdio Config | MCP 标准配置

ModelScope/Smithery style (command: npx, args: [package]):

{
  "mcpServers": {
    "metaso-search-mcp": {
      "command": "npx",
      "args": ["metaso-search-mcp"],
      "env": {
        "METASO_API_KEY": "mk-你的API密钥"
      }
    }
  }
}

Local deployment (built from source, run dist/index.js) template:

{
  "mcpServers": {
    "metaso-search-mcp-local": {
      "command": "node",
      "args": ["E:\\\\Development\\\\metasota-API-MCP\\\\dist\\\\index.js"],
      "env": {
        "METASO_API_KEY": "mk-your-api-key"
      }
    }
  }
}

macOS/Linux path example:

{
  "mcpServers": {
    "metaso-search-mcp-local": {
      "command": "node",
      "args": ["/path/to/metasota-API-MCP/dist/index.js"],
      "env": {
        "METASO_API_KEY": "mk-your-api-key"
      }
    }
  }
}

One-command Config Generator | 一键生成配置

Print config JSON:

npx -y -p metaso-search-mcp metaso-mcp-config --client generic --api-key mk-你的API密钥 --print

Write/merge into a target config file:

npx -y -p metaso-search-mcp metaso-mcp-config --client cherry --api-key mk-你的API密钥 --output ./mcp-config.json

Notes:

  • --output will merge into existing mcpServers instead of overwriting all servers.

  • Default server key is metaso-search-mcp; change with --server-name.

API Key & Env Parameters | API Key 与环境变量

Required | 必填:

  • METASO_API_KEY: Metaso API key / 秘塔 API 密钥

Optional | 可选:

  • METASO_BASE_URL (default: https://metaso.cn/api/v1)

  • METASO_TIMEOUT (default: 60000)

  • METASO_DEFAULT_SEARCH_SCOPE (webpage|document|paper|image|video|podcast)

  • METASO_DEFAULT_SEARCH_SIZE (1-100, default 10)

  • METASO_DEFAULT_INCLUDE_SUMMARY (true|false)

  • METASO_DEFAULT_INCLUDE_RAW_CONTENT (true|false)

  • METASO_DEFAULT_CONCISE_SNIPPET (true|false)

  • METASO_DEFAULT_READER_FORMAT (markdown|json)

Local env setup:

cp .env.example .env
# then edit .env and fill METASO_API_KEY

npm Scripts | 脚本

npm run typecheck
npm run build
npm run rebuild
npm run release:check
npm run pack:dry
npm run pack:local

Auto Publish to npm (GitHub Actions) | 自动发布 npm

Workflow: .github/workflows/release-npm.yml

  1. Add repository secret NPM_TOKEN in GitHub Actions secrets.

  2. Push a version tag:

git tag v1.0.1
git push origin v1.0.1

Platform Publishing Docs | 平台发布文档

See docs/PLATFORM_PUBLISHING.md for:

  • Smithery submission checklist

  • ModelScope (魔塔社区) submission checklist

  • npx stdio config snippet for platform pages

License

MIT

Available Tools

2 tools
metaso_readerA

读取指定URL的网页内容并转换为结构化格式。

支持的输出格式:

  • markdown: 返回Markdown格式的网页内容 (默认)

  • json: 返回JSON格式，包含标题、URL、内容等结构化信息

适用场景:

  • 提取新闻文章内容

  • 获取网页正文

  • 将网页转换为可读格式

ParametersJSON Schema
NameRequiredDescriptionDefault
urlYes要读取的网页URL (必填)
formatNo输出格式markdown

TDQS

A3.8/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It discloses the core behavior (reads URL and converts format) and the output formats, which implies a read-only operation. However, it does not mention potential failure modes, restrictions (e.g., paywall, JavaScript-heavy pages), rate limits, or authentication needs. Since the tool seems simple and read-only, a 3 is appropriate—it adds some context but lacks depth.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise and well-structured: an initial statement, a bulleted list of formats, and a bulleted list of use cases. Every sentence adds value, with no redundant phrases. It is appropriately sized and easy to scan.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool is simple (2 params, no nested objects, no output schema), and the description covers purpose, formats, defaults, and use cases. It lacks explicit notes on error handling or limitations, but given the low complexity and clear schema, the description is largely complete. A 4 acknowledges minor missing behavioral context.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema already covers both parameters with descriptions and the enum for format; schema coverage is 100%. The description adds a list of supported output formats and default, but this largely duplicates the schema's 'default' field. The description does not provide additional meaning beyond the schema, so baseline 3 is fair.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly specifies that the tool reads a given URL's web content and converts it to structured formats (markdown/json). The verb '读取' and resource 'URL的网页内容' are specific, and the mention of structured output distinguishes it from a generic fetch tool. However, it does not explicitly contrast with the sibling tool 'metaso_search' (which likely searches), so it stops short of a 5.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description lists concrete applicable scenarios such as extracting news articles, getting main body text, and converting pages to readable format. This provides clear context for when to use the tool. It does not explicitly exclude cases (e.g., 'do not use for search' or 'only for public pages'), but the scenarios are sufficient for basic guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 2 tool updatesv1.1.2
    • First observedmetaso_reader
    • First observedmetaso_search
Deploy Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A3.9/5.0

Scored across 2 tools

Disambiguation5/5

The two tools have clearly distinct purposes: one performs searches across multiple scopes, and the other reads and converts a specific URL. There is no ambiguity between them.

Naming Consistency5/5

Both tools follow the same 'metaso_' prefix and use underscore-separated lowercase words. The naming pattern is consistent and predictable.

Tool Count3/5

With only two tools, the set is at the low end of acceptable. For a search-and-read server, the minimalism is defensible, but it feels slightly thin compared to a typical well-scoped server.

Completeness4/5

The search tool covers diverse scopes (web, documents, papers, images, videos, podcasts) and the reader handles URL content extraction. Together they cover the core search-and-read pipeline, though advanced features like search history or saved results are missing.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers