concept-rag
Verified Safeby m2ux
Overview
A RAG-based conceptual search system that transforms simple vector search into sophisticated conceptual search for document libraries.
Installation
node dist/conceptual_index.js ~/.concept_ragEnvironment Variables
- OPENROUTER_API_KEY
- DATABASE_URL
- CATALOG_TABLE_NAME
- CHUNKS_TABLE_NAME
- CONCEPTS_TABLE_NAME
- CATEGORIES_TABLE_NAME
- OPENROUTER_SUMMARY_MODEL
- OPENROUTER_CONCEPT_MODEL
- EMBEDDING_PROVIDER
- EMBEDDING_MODEL
- EMBEDDING_DIMENSIONS
- EMBEDDING_BATCH_SIZE
- SEARCH_DEFAULT_LIMIT
- SEARCH_MAX_LIMIT
- SEARCH_WEIGHT_VECTOR
- SEARCH_WEIGHT_BM25
- SEARCH_WEIGHT_TITLE
- SEARCH_WEIGHT_CONCEPT
- SEARCH_WEIGHT_WORDNET
- ENABLE_CACHING
- CACHE_TTL_MS
- PRELOAD_CACHES
- LOG_LEVEL
- DEBUG_SEARCH
- ENABLE_TIMING
- DEBUG_OCR
Security Notes
The server utilizes `child_process.spawn` and `execSync` for OCR (Tesseract) and WordNet lookups (Python script). Inputs to these external commands, such as PDF file paths and WordNet query terms, are sanitized for common injection patterns (e.g., quoting file paths, escaping string literals for Python). This mitigates basic command injection risks, as documented in ADRs (e.g., SQL injection prevention). However, sophisticated attack vectors against external process execution with user-derived inputs can be challenging to fully eliminate, requiring careful auditing of all input transformations. No direct `eval` or blatant obfuscation found.
Similar Servers
UltraRAG
A low-code RAG framework for researchers to build and iterate on complex multi-stage, multimodal Retrieval-Augmented Generation (RAG) pipelines using a Model Context Protocol (MCP) architecture.
mcp-local-rag
Provides a local RAG-like web search capability for LLMs through the Model Context Protocol without external APIs.
pageindex-mcp
Provides vectorless, reasoning-based RAG capabilities for LLMs to navigate and retrieve information from hierarchical document structures, primarily for long PDFs.
flexible-graphrag
The Flexible GraphRAG MCP Server provides a Model Context Protocol (MCP) interface for AI assistants (like Claude Desktop) to interact with a sophisticated RAG and GraphRAG system for document processing, knowledge graph auto-building, hybrid search, and AI Q&A.