Simple Files Vectorstore
Provides semantic search across local files by creating vector embeddings from watched directories.
@lishenxydlgzs/simple-files-vectorstore
A Model Context Protocol (MCP) server that provides semantic search capabilities across files. This server watches specified directories and creates vector embeddings of file contents, enabling semantic search across your documents.
Installation & Usage
Add to your MCP settings file:
{
"mcpServers": {
"files-vectorstore": {
"command": "npx",
"args": [
"-y",
"@lishenxydlgzs/simple-files-vectorstore"
],
"env": {
"WATCH_DIRECTORIES": "/path/to/your/directories"
},
"disabled": false,
"autoApprove": []
}
}
}
MCP settings file locations:
- VSCode Cline Extension:
~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json - Claude Desktop App:
~/Library/Application Support/Claude/claude_desktop_config.json
Configuration
The server requires configuration through environment variables:
Required Environment Variables
You must specify directories to watch using ONE of the following methods:
WATCH_DIRECTORIES: Comma-separated list of directories to watchWATCH_CONFIG_FILE: Path to a JSON configuration file with awatchListarray
Example using WATCH_DIRECTORIES:
{
"mcpServers": {
"files-vectorstore": {
"command": "npx",
"args": [
"-y",
"@lishenxydlgzs/simple-files-vectorstore"
],
"env": {
"WATCH_DIRECTORIES": "/path/to/dir1,/path/to/dir2"
},
"disabled": false,
"autoApprove": []
}
}
}
Example using WATCH_CONFIG_FILE:
{
"mcpServers": {
"files-vectorstore": {
"command": "npx",
"args": [
"-y",
"@lishenxydlgzs/simple-files-vectorstore"
],
"env": {
"WATCH_CONFIG_FILE": "/path/to/watch-config.json"
},
"disabled": false,
"autoApprove": []
}
}
}
The watch config file should have the following structure:
{
"watchList": [
"/path/to/dir1",
"/path/to/dir2",
"/path/to/specific/file.txt"
]
}
Optional Environment Variables
CHUNK_SIZE: Size of text chunks for processing (default: 1000)CHUNK_OVERLAP: Overlap between chunks (default: 200)IGNORE_FILE: Path to a .gitignore style file to exclude files/directories based on patterns
Example with all optional parameters:
{
"mcpServers": {
"files-vectorstore": {
"command": "npx",
"args": [
"-y",
"@lishenxydlgzs/simple-files-vectorstore"
],
"env": {
"WATCH_DIRECTORIES": "/path/to/dir1,/path/to/dir2",
"CHUNK_SIZE": "2000",
"CHUNK_OVERLAP": "500",
"IGNORE_FILE": "/path/to/.gitignore"
},
"disabled": false,
"autoApprove": []
}
}
}
MCP Tools
This server provides the following MCP tools:
1. search
Perform semantic search across indexed files.
Parameters:
query(required): The search query stringlimit(optional): Maximum number of results to return (default: 5, max: 20)
Example response:
[
{
"content": "matched text content",
"source": "/path/to/file",
"fileType": "markdown",
"score": 0.85
}
]
2. get_stats
Get statistics about indexed files.
Parameters: None
Example response:
{
"totalDocuments": 42,
"watchedDirectories": ["/path/to/docs"],
"processingFiles": []
}
Features
- Real-time file watching and indexing
- Semantic search using vector embeddings
- Support for multiple file types
- Configurable chunk size and overlap
- Background processing of files
- Automatic handling of file changes and deletions
Repository
相關伺服器
ContextMCP
A self-hosted MCP server that indexes documentation from various sources and serves it to AI Agents with semantic search.
Not Human Search
MCP search engine for agent-native services. Find 8,600+ APIs, MCP servers, and agentic sites ranked by agentic readiness score (llms.txt, OpenAPI, ai-plugin, MCP) - live MCP endpoint at nothumansearch.ai/mcp with 8 tools.
RedNote MCP
Search and retrieve content from the Xiaohongshu (Red Book) platform.
ProPublica MCP Server
Search and analyze nonprofit organizations' Form 990 data using ProPublica's Nonprofit Explorer API.
EzBiz Government Contracting Intelligence
AI-powered federal contract search, agency spending analysis, competitor win tracking, and set-aside monitoring.
Jewish Library MCP Server
A full-text search server for Jewish texts and literature.
Yandex Search
A web search server that uses the Yandex Search API.
鸣潮 MCP Server
Retrieves character and artifact information from the game Wuthering Waves (鸣潮) as Markdown, optimized for large language models.
BGPT MCP API
Search scientific papers from any MCP tool. Raw experimental data from full-text papers — methods, results, quality scores. 50 free searches, then $0.01/result.
Google Search Engine
A server for Google search and webpage content extraction, built on Cloudflare Workers with OAuth support.