Dolphin MCP

Dolphin MCP

Provides AI-powered semantic code search across multiple repositories, allowing natural language queries to find code chunks and retrieve specific file contents through Dolphin AI embeddings.

Category
访问服务器

README

dolphin-mcp

NPM Version License: MIT

MCP server for Dolphin semantic code search. Conforms to MCP spec.

Quick Start

No installation needed - use bunx:

bunx dolphin-mcp

Configuration

Continue.dev

Add to config.yaml:

mcpServers:
  - name: Dolphin-KB
    command: bunx
    args:
      - dolphin-mcp
    env:
      DOLPHIN_API_URL: "http://127.0.0.1:7777"
      # Optional: Performance optimization for parallel snippet fetching
      MAX_CONCURRENT_SNIPPET_FETCH: "8"
      SNIPPET_FETCH_TIMEOUT_MS: "2000"
      SNIPPET_FETCH_RETRY_ATTEMPTS: "1"

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "dolphin-kb": {
      "command": "bunx",
      "args": ["dolphin-mcp"],
      "env": {
        "DOLPHIN_API_URL": "http://127.0.0.1:7777",
        "MAX_CONCURRENT_SNIPPET_FETCH": "8",
        "SNIPPET_FETCH_TIMEOUT_MS": "2000",
        "SNIPPET_FETCH_RETRY_ATTEMPTS": "1"
      }
    }
  }
}

Environment Variables

Variable Default Description
DOLPHIN_API_URL http://127.0.0.1:7777 Dolphin API endpoint
LOG_LEVEL info Logging level (debug, info, warn, error)

Parallel Snippet Fetching Configuration

These variables control the performance optimization for parallel snippet fetching in search_knowledge:

Variable Default Description Recommended Range
MAX_CONCURRENT_SNIPPET_FETCH 8 Maximum parallel snippet requests 4-12
SNIPPET_FETCH_TIMEOUT_MS 2000 Timeout per snippet request (ms) 1500-3000
SNIPPET_FETCH_RETRY_ATTEMPTS 1 Retry attempts for failed requests 0-3

Configuration Presets

Conservative (recommended for limited resources):

MAX_CONCURRENT_SNIPPET_FETCH=4
SNIPPET_FETCH_TIMEOUT_MS=1500
SNIPPET_FETCH_RETRY_ATTEMPTS=1

Recommended (balanced performance):

MAX_CONCURRENT_SNIPPET_FETCH=8
SNIPPET_FETCH_TIMEOUT_MS=2000
SNIPPET_FETCH_RETRY_ATTEMPTS=1

Performance (maximum throughput):

MAX_CONCURRENT_SNIPPET_FETCH=10
SNIPPET_FETCH_TIMEOUT_MS=3000
SNIPPET_FETCH_RETRY_ATTEMPTS=2

Available Tools

search_knowledge

Semantically query code and docs across indexed repositories and return ranked snippets with citations.

{
  "query": "string (required)",
  "repos": ["string"],
  "path_prefix": ["string"],
  "top_k": "number (1-100)",
  "max_snippets": "number",
  "embed_model": "small | large",
  "score_cutoff": "number"
}

fetch_chunk

Fetch a chunk by chunk_id and return fenced code with citation.

{
  "chunk_id": "string (required)"
}

fetch_lines

Fetch a file slice [start, end] inclusive from disk and return fenced code with citation.

{
  "repo": "string (required)",
  "path": "string (required)",
  "start": "number (required, 1-indexed)",
  "end": "number (required, inclusive)"
}

get_vector_store_info

Report namespaces, dims, limits, and approximate counts.

{}

open_in_editor

Compute a vscode://file URI for a repo path and optional position.

{
  "repo": "string (required)",
  "path": "string (required)",
  "line": "number (1-indexed)",
  "column": "number (1-indexed)"
}

Installation (Optional)

If you prefer installing globally:

bun install -g dolphin-mcp

Then use dolphin-mcp instead of bunx dolphin-mcp.

Requirements

  • Bun >= 1.0.0 - Install
  • Dolphin API running on configured endpoint

License

MIT - see LICENSE file for details.

Links

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选