Skip to main content
Glama
kh-54

Jina MCP Server

by kh-54
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Hybrid

Jina MCP Server

A lightweight MCP server that exposes web_reader via r.jina.ai.

  • stdio — for 1Panel / Cursor local (default)

  • Streamable HTTP — optional standalone remote mode (node http.js)

Tool: web_reader

{ "url": "https://example.com" }

Related MCP server: Scrapi MCP Server

1Panel 部署（推荐）

1Panel 会把 stdio MCP 自动转成 SSE 对外暴露，无需自己写 HTTP 服务。

目录规划（建议统一管理所有 MCP）

/opt/mcp-servers/
  jina/          ← 本项目
  fetch/         ← 以后的 MCP
  github/        ← 以后的 MCP

第一次：部署 Jina MCP

1. 上传代码到服务器

mkdir -p /opt/mcp-servers
cd /opt/mcp-servers
git clone <your-repo-url> jina
cd jina
npm install --omit=dev

2. 在 1Panel 创建 MCP Server

进入 AI → MCP → 创建 MCP Server，选择 二进制方式运行：

配置项

填写

名称

jina

启动命令

node

启动参数

/opt/mcp-servers/jina/stdio.js

挂载目录

宿主机 /opt/mcp-servers/jina → 容器 /opt/mcp-servers/jina

如果面板只有 npx 方式、且你的 MCP 已发布到 npm，也可以用： 命令 npx，参数 -y jina-mcp-server

3. 绑定网站（对外访问）

在 MCP 详情里绑定域名，例如：

  • 域名：mcp.example.com

  • SSE 路径：/jina/sse（每个 MCP 用不同路径）

多个 MCP 可共用同一域名：

https://mcp.example.com/jina/sse
https://mcp.example.com/fetch/sse
https://mcp.example.com/github/sse

4. 安全配置

  • 开启 HTTPS（Let's Encrypt 或上传证书）

  • 配置 IP 白名单（限制可访问的客户端 IP）

5. 获取客户端配置

点击 MCP 实例的 「配置」 按钮，复制生成的 JSON 到 Cursor：

{
  "mcpServers": {
    "jina": {
      "url": "https://mcp.example.com/jina/sse"
    }
  }
}

以后新增 MCP 的固定流程

每个新 MCP 重复以下步骤即可：

# 1. 放到统一目录
cd /opt/mcp-servers
git clone <new-mcp-repo> <name>
cd <name> && npm install --omit=dev   # 或 pip install / go build 等

# 2. 1Panel → AI → MCP → 创建
#    命令: node (或 npx / python / 二进制)
#    参数: /opt/mcp-servers/<name>/stdio.js
#    挂载: /opt/mcp-servers/<name>

# 3. 绑定同一域名，换不同 SSE 路径
#    例如 /fetch/sse、/github/sse

# 4. 复制客户端配置发给用户

MCP 类型

启动命令示例

Node.js（本项目）

node /opt/mcp-servers/jina/stdio.js

npm 包

npx -y @modelcontextprotocol/server-fetch

Python

uvx mcp-server-fetch

编译好的二进制

/opt/mcp-servers/xxx/bin/mcp-xxx

本地开发

npm install
npm start              # stdio，给 1Panel / Cursor command 模式用
npm run start:http     # 独立 HTTP 模式（不经 1Panel）

Cursor 本地 stdio 配置

{
  "mcpServers": {
    "jina": {
      "command": "node",
      "args": ["/path/to/jina_mcp_server/stdio.js"]
    }
  }
}

独立 HTTP 模式（不用 1Panel 时）

MCP_PORT=8000 node http.js
{
  "mcpServers": {
    "jina": {
      "url": "https://mcp.example.com/mcp",
      "headers": {
        "Authorization": "Bearer ${env:JINA_MCP_TOKEN}"
      }
    }
  }
}

变量

默认

说明

MCP_PORT

8000

HTTP 端口

MCP_HOST

0.0.0.0

绑定地址

MCP_AUTH_TOKEN

—

Bearer 鉴权

MCP_ALLOWED_HOSTS

—

反代 Host 白名单

Docker（可选，非 1Panel 场景）

docker compose up -d --build

Available Tools

1 tool
web_readerC

Fetch clean webpage content via Jina Reader

ParametersJSON Schema
NameRequiredDescriptionDefault
urlYesURL to fetch

TDQS

C2.9/5.0
Behavior2/5

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

No annotations exist, so the description carries full burden. It does not disclose any behavioral traits such as handling of dynamic content, error responses, rate limits, or authentication requirements.

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 a single, front-loaded sentence with no wasted words. Every part contributes to the core purpose.

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

Completeness2/5

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

Given the lack of output schema and siblings, the description fails to explain what the tool returns (e.g., format, structure) or any constraints like URL validity or JavaScript rendering. It is incomplete for an agent to use confidently.

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?

Schema coverage is 100% with the 'url' parameter described as 'URL to fetch'. The description adds no additional meaning beyond the schema, so baseline score of 3 is appropriate.

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 states the verb 'fetch' and the resource 'clean webpage content', and mentions the method 'Jina Reader'. Although it lacks detail on what 'clean' means, it effectively communicates the tool's purpose.

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

Usage Guidelines2/5

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

No guidance is provided on when to use this tool vs alternatives. With no sibling tools listed, there is no contextual advice, leaving the agent without usage boundaries.

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. 1 tool updatev1.0.0
    • First observedweb_reader
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

B3/5.0

Scored across 1 tool

Disambiguation5/5

With only one tool, there is no possibility of confusion between tools. The single tool's purpose is clear.

Naming Consistency4/5

The naming is descriptive and follows a common pattern (verb_noun). As a single tool, consistency is not an issue.

Tool Count2/5

Only one tool is provided, which is too few for a server named 'Jina MCP Server'. The server likely expects more tools (e.g., search, summarize) to justify its name and purpose.

Completeness2/5

The server only covers webpage reading, missing other expected functionalities such as searching or processing content. This is a significant gap relative to the server's implied scope.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • MCP Website Fetcher

    electrocucaracha
    F
    license
    Not graded
    quality
    F
    maintenance
    A lightweight MCP server implementation that provides a simple tool for fetching website content, demonstrating how tools can be exposed and interacted with using the MCP protocol.
    -

  • jina-mcp-server

    snoopygao
    A
    license
    Not graded
    quality
    C
    maintenance
    A remote Model Context Protocol (MCP) server providing access to Jina Reader, Embeddings, and Reranker APIs for URL-to-markdown, web search, image search, and embeddings/reranker tools.
    Apache 2.0