milvus-text-search-mcp
MCP Server for semantic text search in Milvus: converts natural language to vectors automatically — no manual vector input required.
将自然语言查询文本自动转换为向量嵌入,再执行 Milvus 向量相似度搜索。
无需 Agent 手动传入 1024 维向量数组,解决了 mcp-server-milvus 的 milvus-vector-search 工具需要传入向量的限制。
安装
# 使用 uvx 直接运行(无需预装,推荐)
uvx milvus-text-search-mcp
# 或使用 uv tool install 预装
uv tool install milvus-text-search-mcp
# 或使用 pip
pip install milvus-text-search-mcp
配置(环境变量)
| 变量 | 说明 | 默认值 |
|---|---|---|
MILVUS_URI |
Milvus 服务地址 | http://localhost:19530 |
MILVUS_USER |
Milvus 认证用户名(启用鉴权时必填) | 无 |
MILVUS_PASSWORD |
Milvus 认证密码(启用鉴权时必填) | 无 |
MILVUS_DB |
指定连接的数据库名称(留空则使用 default) |
无 |
MILVUS_COLLECTION |
要搜索/统计的集合名称(必填) | 无 |
EMBEDDING_API_BASE |
OpenAI 兼容 Embedding API 基础地址 | 无 |
EMBEDDING_API_KEY |
Embedding API 密钥(必填) | 无 |
EMBEDDING_MODEL |
Embedding 模型名称(必填) | 无 |
支持任何 OpenAI 兼容的 Embedding API,如 SiliconFlow、Azure OpenAI、本地部署的 Ollama 等。
在 workmate 平台配置工具集
在工具集配置 JSON 中填写(替换为你的实际地址和密钥):
{
"command": "uvx",
"args": ["milvus-text-search-mcp"],
"env": {
"MILVUS_URI": "http://your-milvus-host:19530",
"MILVUS_USER": "your-milvus-username",
"MILVUS_PASSWORD": "your-milvus-password",
"MILVUS_DB": "your-milvus-db",
"MILVUS_COLLECTION": "resumes_collection",
"EMBEDDING_API_BASE": "https://your-embedding-api/v1",
"EMBEDDING_API_KEY": "your-api-key",
"EMBEDDING_MODEL": "your-embedding-model-name"
},
"timeout": 120
}
提供的工具
milvus_semantic_text_search(核心工具)
将自然语言文本转换为向量后执行语义相似度搜索。
# Agent 调用示例
# 注意:collection 由环境变量 MILVUS_COLLECTION 决定,不是调用参数
milvus_semantic_text_search(
query_text="有5年Java开发经验的后端工程师",
top_k=10,
filter_expr='age >= 25 && age <= 40', # 可选
output_fields="name,age,email" # 可选
)
milvus_count_records(统计工具)
统计集合中满足条件的记录总数,专为"有多少条"、"上传了多少"等数量查询设计。
# Agent 调用示例
milvus_count_records(
filter_expr='resume_upload_time >= "2026-06-10T00:00:00+00:00"' # 可选
)
# 返回:{"count": 87, "collection": "resumes_collection", "filter": "..."}
⚠️ 不要用
mcp-server-milvus的milvus_query+count(*)统计数量,该工具内部强制limit,与count(*)不兼容,会报错。请始终使用此工具统计数量。
milvus_get_text_embedding(辅助工具)
仅将文本转换为向量嵌入,不执行搜索。
通常不需要直接调用,请优先使用 milvus_semantic_text_search。
与 mcp-server-milvus 的关系
| 工具集 | 输入 | 适用场景 |
|---|---|---|
mcp-server-milvus |
需要 float 向量数组 | 已有向量的场景 |
milvus-text-search-mcp(本包) |
自然语言文本 | 语义搜索场景(推荐) |
两者可同时配置给同一个数字员工,互补使用。
License
MIT
Metadata
Release files for milvus-text-search-mcp 0.7.0
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 |
|---|---|---|---|---|
| milvus_text_search_mcp-0.7.0.tar.gz | 91.5 kB | Without provenance | Details for milvus_text_search_mcp-0.7.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| milvus_text_search_mcp-0.7.0-py3-none-any.whl | Python 3 | none | any | Details for milvus_text_search_mcp-0.7.0-py3-none-any.whl |
Total release size: 100.4 kB
Release files / milvus_text_search_mcp-0.7.0.tar.gz
| Download URL | milvus_text_search_mcp-0.7.0.tar.gz |
|---|---|
| Size | 91.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
877b8cf62cb27bd0b90212ad8564016d4a642f2be7bc887dc35348d8ee93e69f
|
|
BLAKE2b-256 checksum How to use checksums |
ecce57557d87bd7156ca7602f1948bb232a68f93e0d3c96d85e6f4dec5c23e38
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.14 {"installer":{"name":"uv","version":"0.11.14","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":null,"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / milvus_text_search_mcp-0.7.0-py3-none-any.whl
| Download URL | milvus_text_search_mcp-0.7.0-py3-none-any.whl |
|---|---|
| Size | 8.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
555ff96402aa246d1c367218a6a3eca57c0ac92d6672710b3f2b5ce077d29b50
|
|
BLAKE2b-256 checksum How to use checksums |
7abcb6e3de3124a57987eb6a3ff15bd12756516c4b457d7a6da5ea676a331e8f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.14 {"installer":{"name":"uv","version":"0.11.14","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":null,"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for milvus-text-search-mcp 0.7.0
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.