Brave Search MCP Server
Brave Search MCP 服务器
该项目为 Brave Search 实现了模型上下文协议 (MCP) 服务器,允许与 Claude 等 AI 助手集成。
先决条件
Python 3.11+
uv - 快速的 Python 软件包安装程序和解析器
Related MCP server: Brave Search MCP Server
安装
通过 Smithery 安装
要通过Smithery自动为 Claude Desktop 安装 Brave Search MCP 服务器:
npx -y @smithery/cli install @arben-adm/brave-mcp-search --client claude手动安装
克隆存储库:
git clone https://github.com/your-username/brave-search-mcp.git cd brave-search-mcp使用 uv 创建虚拟环境并安装依赖项:
uv venv source .venv/bin/activate # On Windows, use: .venv\Scripts\activate uv pip install -r requirements.txt设置您的 Brave Search API 密钥:
export BRAVE_API_KEY=your_api_key_here在 Windows 上,使用:
set BRAVE_API_KEY=your_api_key_here
用法
配置您的 MCP 设置文件(例如,
claude_desktop_config.json)以包含 Brave Search MCP 服务器:{ "mcpServers": { "brave-search": { "command": "uv", "args": [ "--directory", "path-to\\mcp-python\\brave-mcp-search\\src", "run", "server.py" ], "env": { "BRAVE_API_KEY": "YOUR_BRAVE_API_KEY_HERE" } } } }用您的实际 Brave API 密钥替换
YOUR_BRAVE_API_KEY_HERE。通过运行具有更新配置的 MCP 兼容 AI 助手来启动 Brave Search MCP 服务器。
服务器现在将运行并准备接受来自 MCP 客户端的请求。
您现在可以通过调用可用的工具在与 MCP 兼容的 AI 助手(如 Claude)中使用 Brave Search 功能。
可用工具
该服务器提供两个主要工具:
brave_web_search:使用 Brave Search API 执行网络搜索。brave_local_search:搜索本地企业和地点。
有关详细使用信息,请参阅src/server.py中的工具文档字符串。
发展
要对项目进行更改:
根据需要修改
src目录中的代码。如果添加或删除依赖项,请更新
requirements.txt文件:uv pip freeze > requirements.txt重新启动服务器以应用更改。
故障排除
如果您遇到任何问题:
确保您的 Brave API 密钥已正确设置。
检查所有依赖项是否已安装。
验证您使用的 Python 版本是否兼容。
如果您对代码进行了更改,请确保重新启动服务器。
执照
该项目根据 MIT 许可证获得许可 - 有关详细信息,请参阅LICENSE文件。
Available Tools
2 toolsbrave_local_searchC
Search for local businesses and places
Args:
query: Location terms
count: Results (1-20
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| count | No |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. While 'Search' implies a read-only operation, it doesn't specify authentication requirements, rate limits, data sources, or what happens when no results are found. The description provides minimal behavioral context beyond the basic operation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately concise with a clear purpose statement followed by parameter explanations. The formatting with 'Args:' section is helpful, though the incomplete 'Results (1-20' text suggests a typographical error that slightly detracts from professionalism.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a search tool with 2 parameters, 0% schema description coverage, no annotations, and no output schema, the description is insufficient. It doesn't explain what information is returned, result format, error conditions, or how this differs from the sibling web search tool, leaving significant gaps for an AI agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It provides basic explanations for both parameters ('Location terms' for query, 'Results (1-20' for count), but these are minimal and don't explain format expectations, what constitutes valid 'Location terms', or how the count parameter works with the default value of 20.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose as 'Search for local businesses and places' which is a specific verb+resource combination. However, it doesn't explicitly differentiate from its sibling 'brave_web_search' which likely searches the broader web rather than local businesses.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
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 its sibling 'brave_web_search' or any alternatives. There's no mention of appropriate contexts, prerequisites, or exclusions for using this local search functionality.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
brave_web_searchC
Execute web search using Brave Search API with improved results
Args:
query: Search terms
count: Desired number of results (10-20)
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| count | No |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions 'improved results' but doesn't explain what this means (e.g., better ranking, additional metadata). It lacks details on rate limits, authentication needs, error handling, or what the output looks like (e.g., format of search results). For a tool with no annotation coverage, this is a significant gap in transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and front-loaded, starting with the core purpose. The two-sentence structure is efficient, with the second sentence dedicated to parameter details. There's minimal waste, though the formatting with indentation and quotes might slightly affect readability in some contexts.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (a web search tool with 2 parameters), lack of annotations, and no output schema, the description is incomplete. It doesn't explain the return values (e.g., what 'improved results' include), error conditions, or usage constraints. For a tool that interacts with an external API, more context is needed to ensure reliable agent invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It adds basic semantics for both parameters: 'query' as 'Search terms' and 'count' as 'Desired number of results (10-20)'. This clarifies the purpose of each parameter beyond the schema's titles ('Query' and 'Count'). However, it doesn't provide deeper context like query syntax examples or why count has a range, leaving room for improvement.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Execute web search using Brave Search API with improved results.' It specifies the verb ('Execute web search') and resource ('Brave Search API'), making it easy to understand what the tool does. However, it doesn't explicitly differentiate from its sibling 'brave_local_search' (e.g., by mentioning this is for general web searches vs. local searches), which prevents 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.
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. It doesn't mention the sibling tool 'brave_local_search' or any other potential alternatives, nor does it specify contexts or prerequisites for usage. The only implied usage is for web searches, but this is too vague for effective tool selection.
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.
2 tool updates
v1.0.0- First observed
brave_local_search - First observed
brave_web_search
TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: one is for local business/place searches and the other is for general web searches. The descriptions explicitly differentiate their domains (local vs. web), making it impossible to confuse them. Each tool serves a unique function within the search domain.
Both tools follow a perfectly consistent naming pattern: 'brave_' prefix followed by descriptive snake_case (local_search, web_search). The naming convention is uniform across all tools, making them predictable and easy to understand. There are no deviations or mixed styles.
With only 2 tools, the server feels somewhat thin for a search domain that could benefit from more specialized operations (e.g., image search, news search, autocomplete). While the two core search functions are covered, the limited tool count may restrict agent capabilities for broader search-related tasks. It's borderline minimal but functional.
The server covers the essential search operations (local and web) well, but there are minor gaps in the search surface. For instance, no tools for image search, news search, or search suggestions/autocomplete are included, which are common in search APIs. However, the core workflows are adequately supported, and agents can work around these omissions.
Maintenance
Related MCP Connectors
Visit https://brave.com/search/api/ for a free API key. Search the web, local businesses, images,…
Provides AI assistants with access to Seltz's powerful Web Search capabilities.
Web search, browser automation, scraping, crawling and CAPTCHA solving for AI agents.
Brave Search MCP — independent web index (no Google/Bing dependency)
Related MCP Servers
- AlicenseAqualityCmaintenanceA Model Context Protocol server that provides real-time web search capabilities to AI assistants through pluggable search providers, currently integrated with the Brave Search API.516MIT
- AlicenseNot gradedqualityNot gradedmaintenanceIntegrates Brave Search API to enable web, image, video, news, and local business searches with filtering options like result count and content freshness.399 npm-
- AlicenseNot gradedqualityDmaintenanceIntegrates the Brave Search API to provide web and local search capabilities, including news, articles, and business information. It features smart fallbacks from local to web search and supports pagination and freshness controls.399 npm1GPL 3.0
- AlicenseNot gradedqualityDmaintenanceIntegrates the Brave Search API into AI assistants to enable web and local business search capabilities. This allows models to perform real-time information retrieval and locate places using the Model Context Protocol.1MIT