Schema | RAG-Anything MCP Server

RAG-Anything MCP Server

Describes the environment variables required to run the server.

Name	Required	Description	Default
`LLM_MODEL`	No	Language model to use for LLM tasks (default: gpt-4o-mini). Overridden by MODEL_PROFILE.	gpt-4o-mini
`LOG_LEVEL`	No	Logging level (default: INFO).	INFO
`UPLOAD_DIR`	No	Directory for uploaded files (default: ./uploads).	./uploads
`VISION_MODEL`	No	Vision model to use (default: gpt-4o). Overridden by MODEL_PROFILE.	gpt-4o
`EMBEDDING_DIM`	No	Dimension of embeddings (default: 3072). Overridden by MODEL_PROFILE.	3072
`MAX_FILE_SIZE`	No	Maximum file size in MB (default: 100).	100
`MODEL_PROFILE`	No	Name of a model profile defined in MODEL_PROFILES or MODEL_PROFILES_PATH. When set, overrides LLM_MODEL, VISION_MODEL, EMBEDDING_MODEL, and EMBEDDING_DIM.
`MODEL_PROFILES`	No	JSON string containing profile definitions. Used if MODEL_PROFILE is set. Alternative to MODEL_PROFILES_PATH.
`OPENAI_API_KEY`	Yes	Your OpenAI API key. Required.
`EMBEDDING_MODEL`	No	Embedding model to use (default: text-embedding-3-large). Overridden by MODEL_PROFILE.	text-embedding-3-large
`OPENAI_BASE_URL`	No	Optional base URL for OpenAI API (e.g., https://api.openai.com/v1).
`RAG_STORAGE_DIR`	No	Directory for RAG storage (default: ./rag_storage).	./rag_storage
`MODEL_PROFILES_PATH`	No	Path to a JSON file with model profiles (default: ./model_profiles.json). Used if MODEL_PROFILE is set.	./model_profiles.json

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.

Server capabilities have not been inspected yet.

Functions exposed to the LLM to take actions

Name	Description
No tools

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

This server cannot be deployed