MCP RAG
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LOG_LEVEL | No | Logging level | info |
| CHUNK_SIZE | No | Text chunk size for processing | 1000 |
| CHUNK_OVERLAP | No | Text chunk overlap size | 200 |
| MCP_SERVER_NAME | No | MCP server name | mcp-rag |
| MAX_FILE_SIZE_MB | No | Maximum file size in MB | 50 |
| HUGGINGFACE_MODEL | No | Hugging Face model for embeddings | sentence-transformers/all-MiniLM-L6-v2 |
| MCP_SERVER_VERSION | No | MCP server version | 1.0.0 |
| EMBEDDING_DIMENSION | No | Embedding vector dimension | 384 |
| HUGGINGFACE_API_KEY | Yes | Your Hugging Face API key | |
| SUPPORTED_EXTENSIONS | No | Supported file extensions | .pdf,.docx,.txt,.md,.csv,.json |
| CHROMA_COLLECTION_FILES | No | ChromaDB collection name for files | rag_files |
| CHROMA_COLLECTION_MEMORY | No | ChromaDB collection name for agent memory | agent_memory |
| CHROMA_PERSIST_DIRECTORY | No | ChromaDB persistence directory | ./data/chroma |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| add_fileC | Add a file to the RAG system for document retrieval |
| search_filesC | Search for relevant documents in the RAG system |
| remove_fileC | Remove a file from the RAG system |
| list_filesB | List all files in the RAG system |
| add_memoryC | Add information to the agent memory system |
| search_memoryC | Search for relevant information in agent memory |
| remove_memoryC | Remove a memory entry from the agent memory system |
| get_memory_by_agentC | Get all memory entries for a specific agent |
| get_statsC | Get statistics about the RAG system |
| clear_dataC | Clear data from the RAG system |
| test_connectionB | Test the connection to the embedding service |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| RAG Statistics | Current statistics of the RAG system |
TDQS
Scored across 11 tools
Most tools have distinct purposes, but there is some overlap between 'add_file' and 'add_memory' as both involve adding data, though to different systems (RAG vs. memory). Similarly, 'search_files' and 'search_memory' are distinct but conceptually similar operations. The descriptions help clarify the boundaries, but an agent might occasionally confuse these pairs.
All tool names follow a consistent verb_noun pattern with snake_case, such as 'add_file', 'clear_data', 'get_stats', and 'search_memory'. This uniformity makes the set predictable and easy to understand, with no deviations in style or convention.
With 11 tools, the count is well-scoped for a RAG and memory management system. It covers core operations like adding, removing, listing, and searching for both files and memory, along with utility functions like clearing data and testing connections, without being excessive or sparse.
The tool set provides comprehensive coverage for RAG and memory operations, including CRUD-like actions for files and memory entries, search capabilities, and system management. A minor gap is the lack of tools for updating existing files or memory entries, which might require workarounds, but core workflows are well-supported.