Skip to main content
Glama
funtuan

Dify External Knowledge MCP Server

by funtuan
OverviewSchemaRelated ServersScoreDiscussions
JavaScript
Remote

Dify External Knowledge MCP Server

一個 Model Context Protocol (MCP) 伺服器，用於整合 Dify 外部知識庫 API，讓 AI Agent 能夠輕鬆查詢知識庫內容。

安裝

使用 npx (推薦)

npx @funtuantw/dify-external-knowledge-mcp

全域安裝

npm install -g @funtuantw/dify-external-knowledge-mcp

從源碼構建

git clone https://github.com/funtuantw/dify-external-knowledge-mcp.git
cd dify-external-knowledge-mcp
npm install
npm run build

Related MCP server: Candlekeep

功能特性

  • 🔍 查詢 Dify 外部知識庫

  • 📊 支援相關性分數閾值設定

  • 🎯 支援元數據條件篩選

  • 🔐 使用環境變數管理 API 憑證

  • 📦 支援 npx 直接運行

  • 🌐 知識庫 ID 透過環境變數配置，無需每次指定

配置

{
  "mcpServers": {
    "dify-knowledge": {
      "command": "npx",
      "args": ["@funtuantw/dify-external-knowledge-mcp"],
      "env": {
        "DIFY_KNOWLEDGE_ENDPOINT": "https://your-endpoint.com",
        "DIFY_API_KEY": "your-api-key-here",
        "DIFY_KNOWLEDGE_ID": "your-knowledge-id-here"
      }
    }
  }
}

或者使用本地路徑（開發時）：

{
  "mcpServers": {
    "dify-knowledge": {
      "command": "node",
      "args": ["/absolute/path/to/dify-external-knowledge-mcp/dist/index.js"],
      "env": {
        "DIFY_KNOWLEDGE_ENDPOINT": "https://your-endpoint.com",
        "DIFY_API_KEY": "your-api-key-here",
        "DIFY_KNOWLEDGE_ID": "your-knowledge-id-here"
      }
    }
  }
}

使用命令列參數：

{
  "mcpServers": {
    "dify-knowledge": {
      "command": "npx",
      "args": [
        "@funtuantw/dify-external-knowledge-mcp",
        "--endpoint",
        "https://your-endpoint.com",
        "--api-key",
        "your-api-key-here",
        "--knowledge-id",
        "your-knowledge-id-here"
      ]
    }
  }
}

使用方法

可用工具

query_dify_knowledge

查詢 Dify 外部知識庫以檢索相關資訊。

參數：

  • query (必需, string): 用戶的搜尋查詢或問題

  • top_k (可選, number): 返回的最大結果數量（預設：5）

  • score_threshold (可選, number): 最小相關性分數閾值，範圍 0-1（預設：0.5）

  • metadata_condition (可選, object): 元數據篩選條件

    • logical_operator (可選, string): "and" 或 "or"（預設：and）

    • conditions (必需, array): 條件列表

      • name (必需, string[]): 要篩選的元數據欄位名稱

      • comparison_operator (必需, string): 比較運算符（contains, is, not contains, etc.）

      • value (可選, string): 比較值

範例使用（通過 AI Agent）：

請使用 query_dify_knowledge 工具查詢 "什麼是 Dify？"

範例輸出：

Found 2 result(s) for query: "什麼是 Dify？"

### Result 1 (Score: 0.980)
**Title**: knowledge.txt
**Content**:
Dify 是一個 LLMOps 平台，提供 GenAI 應用程序的創新引擎。

**Metadata**: {
  "path": "s3://dify/knowledge.txt",
  "description": "dify 知識文檔"
}
---

### Result 2 (Score: 0.660)
**Title**: introduce.txt
**Content**:
GenAI 應用程序的創新引擎

**Metadata**: {
  "path": "s3://dify/introduce.txt",
  "description": "dify 介紹"
}
---

開發

開發模式運行

npm run dev

構建專案

npm run build

運行構建版本

npm start

API 參考

本專案使用 Dify 外部知識庫 API。

支援的比較運算符

  • contains: 包含某個值

  • not contains: 不包含某個值

  • start with: 以某個值開頭

  • end with: 以某個值結尾

  • is: 等於某個值

  • is not: 不等於某個值

  • empty: 為空

  • not empty: 不為空

  • =: 等於

  • ≠: 不等於

  • >: 大於

  • <: 小於

  • ≥: 大於等於

  • ≤: 小於等於

  • before: 在某個日期之前

  • after: 在某個日期之後

錯誤處理

伺服器會處理以下錯誤情況：

  • 1001: 無效的 Authorization 頭格式

  • 1002: 授權失敗

  • 2001: 知識庫不存在

  • HTTP 403: 訪問被拒絕

  • HTTP 500: 內部伺服器錯誤

專案結構

dify-external-knowledge-mcp/
├── src/
│   └── index.ts          # MCP 伺服器主程式
├── dist/                 # 編譯後的輸出
├── .env.example          # 環境變數範例
├── .gitignore
├── package.json
├── tsconfig.json
├── plan.md              # 專案計畫文檔
└── README.md

授權

MIT

貢獻

歡迎提交 Issue 和 Pull Request！

相關連結

Available Tools

1 tool
query_dify_knowledgeC

Query Dify external knowledge base to retrieve relevant information based on a search query. Returns ranked documents with content, scores, titles, and metadata.

ParametersJSON Schema
NameRequiredDescriptionDefault
queryYesUser's search query or question
top_kNoMaximum number of results to return (default: 5)
score_thresholdNoMinimum relevance score threshold (0-1, default: 0.5)
metadata_conditionNoOptional metadata filtering conditions

TDQS

C2.9/5.0
Behavior2/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions that the tool 'returns ranked documents with content, scores, titles, and metadata,' which gives some output context, but fails to address critical behavioral aspects like rate limits, authentication requirements, error conditions, or whether it's a read-only operation. For a query tool with complex parameters, this leaves significant gaps.

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

Conciseness4/5

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

The description is efficiently structured in two sentences: the first states the purpose, and the second describes the return format. There's no redundant information, and it's front-loaded with the core functionality. A perfect score is withheld because the second sentence could be slightly more concise by integrating return details with the purpose statement.

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

Completeness3/5

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

Given the tool's complexity (4 parameters with nested objects) and lack of annotations or output schema, the description is moderately complete. It covers the basic purpose and return format but misses behavioral details like error handling, performance characteristics, or examples. For a query tool with metadata filtering, more context would be beneficial to fully guide the agent.

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 description does not add any parameter-specific information beyond what's already documented in the input schema, which has 100% coverage. It mentions 'search query' generally but doesn't clarify parameter interactions or usage nuances. With high schema coverage, the baseline score of 3 is appropriate as the schema handles the heavy lifting.

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 tool's purpose: 'Query Dify external knowledge base to retrieve relevant information based on a search query.' It specifies the verb (query), resource (Dify external knowledge base), and action (retrieve information). However, without sibling tools for comparison, it cannot demonstrate differentiation from alternatives, preventing a perfect score.

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?

The description provides no guidance on when to use this tool versus alternatives, prerequisites, or specific contexts. It simply states what the tool does without indicating scenarios where it's appropriate or inappropriate, leaving the agent with minimal usage direction.

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. Dates show when Glama detected each change.

  1. 1 tool updatev1.0.1
    • First observedquery_dify_knowledge
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

B3.1/5.0
Disambiguation5/5

With only one tool, there is no possibility of ambiguity or overlap between tools. The tool's purpose is clearly defined as querying a knowledge base, and no other tools exist to cause confusion.

Naming Consistency5/5

Since there is only one tool, naming consistency is inherently perfect. The tool name 'query_dify_knowledge' follows a clear verb_noun pattern, and there are no other tools to compare it against for inconsistency.

Tool Count2/5

A single tool is too few for a server that appears to handle external knowledge base operations, as it suggests limited functionality. Typically, such a domain would benefit from additional tools like create, update, or delete operations for knowledge management, making this feel incomplete.

Completeness2/5

The tool set is severely incomplete for a knowledge base server. While querying is a core function, there are obvious gaps such as tools for adding, updating, deleting, or listing knowledge base entries, which are essential for full lifecycle coverage in this domain.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • Dify Knowledge MCP Server

    liiklin
    A
    license
    B
    quality
    D
    maintenance
    A Model Context Protocol server that enables AI assistants like Cursor to directly query and retrieve information from Dify knowledge bases through natural language.
    2
    26
    6
    MIT

  • nexacorp-mcp-agent

    phalakmehta
    F
    license
    Not graded
    quality
    C
    maintenance
    A Model Context Protocol server that exposes a hybrid RAG pipeline (dense+sparse retrieval with reranking) for querying an enterprise knowledge base, enabling autonomous agents to search and retrieve relevant information.
    -

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/funtuan/dify-external-knowledge-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server