本地隐私搜索工具
开发者:牛魔科技工作室
零追踪的本地隐私搜索连接器。把 privacy-search 技能的搜索能力封装成 MCP 工具,供 WorkBuddy 及各类 Agent 直接调用。
关键词:开放数据源 API 聚合、Perplexity 式答案合成、本地发起、不留痕、无 Key 可降级。
提供的工具
| 工具 | 作用 | 关键参数 |
|---|---|---|
search |
开放数据源聚合搜索,SimHash 去重 + 多因子加权排序 | query(必填)、engines、num、privacy、no_cache |
synthesize |
Perplexity 式答案合成,基于搜索摘要生成带 citation 的答案 | query(必填)、results、max_sources |
fetch |
抓取开放数据源正文(仅开放域名),三层降级提取 | url(必填)、timeout |
可用引擎
wikipedia、wikidata、arxiv、crossref、openlibrary、pubmed、stackoverflow、github(searxng 需自建,可选)
privacy=strict 时只使用开放数据源(均无需密钥、明确授权),不向任何商业检索服务发送查询词,也不爬取任何第三方网页。
安装
pip install -r requirements.txt
依赖:aiohttp、beautifulsoup4、lxml、jieba、pyyaml。
其中 jieba 是可选依赖:缺失时中文分词自动降级为字符 2-gram,功能完整、排序精度略降,不会报错。
配置(可选)
复制 references/config.yaml.example 到包根目录,改名 config.yaml 后按需修改:
mcp_server.max_results—search默认返回条数(默认 10)mcp_server.max_sources—synthesize默认引用来源数(默认 5)mcp_server.tools— 只想暴露部分工具时填工具名数组,留空表示全部llm.*— 配置后synthesize走真实 LLM 生成;不配置则自动降级为抽取式摘要,无需任何 API Key
不创建 config.yaml 也能直接跑,此时使用 references/config.yaml.example 中的默认值。
隐私说明
- 搜索在本机发起,请求直达各开放数据源的官方 API,不经任何中转服务器
- 结果缓存落在本机,可用
no_cache=true跳过 synthesize若配置了 LLM,摘要才会出域调用模型;不配置则全程本地抽取式摘要
与其它连接器的边界
| 连接器 | 负责 |
|---|---|
| privacy-search | 开放数据源检索、答案合成、开放域名正文提取(信息获取) |
| kingdoc | 金山云端文档的读写与协作(文档生产) |
| wps-office | 本地 Office 文件的排版与转换(文件加工) |
本地调试
python mcp_server.py --schema # 打印工具 schema
python mcp_server.py --test # 协议自测(会真实发起一次搜索)
Metadata
Release files for wb-privacy-search-mcp 1.0.6
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| wb_privacy_search_mcp-1.0.6.tar.gz | 75.9 kB | Without provenance | Details for wb_privacy_search_mcp-1.0.6.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| wb_privacy_search_mcp-1.0.6-py3-none-any.whl | Python 3 | none | any | Details for wb_privacy_search_mcp-1.0.6-py3-none-any.whl |
Total release size: 164.5 kB
Release files / wb_privacy_search_mcp-1.0.6.tar.gz
| Download URL | wb_privacy_search_mcp-1.0.6.tar.gz |
|---|---|
| Size | 75.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ccd68d48416b3d5d064746f5cdc9f522d0c91f395a3da72e1c1536424107763c
|
|
BLAKE2b-256 checksum How to use checksums |
8cc0b03ba6be8f6fa5c89f5f67839d48ca7346be28bc394a88c703899a5273e6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Release files / wb_privacy_search_mcp-1.0.6-py3-none-any.whl
| Download URL | wb_privacy_search_mcp-1.0.6-py3-none-any.whl |
|---|---|
| Size | 88.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
404ee3e9e18628f66f6a8e90c1b46897b7fcd06e9a27cc1e69ce98d644560325
|
|
BLAKE2b-256 checksum How to use checksums |
2e2a09053ac6214e465df4fd5ca7da57daabd2e4fb3217d1dbb5eb0473f73efe
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Security for wb-privacy-search-mcp 1.0.6
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.