Schema | memory-mcp

memory-mcp

Describes the environment variables required to run the server.

Name	Required	Description	Default
`MEMORY_EMBED_DIM`	No	Dimension of the embedding vectors (must match the selected model).	768
`MEMORY_LLM_MODEL`	No	Model name for LLM-based tools (e.g., extract_and_remember, get_trends).	llama3.2
`MEMORY_AI_API_KEY`	No	API key if the provider requires authentication.
`MEMORY_AI_TIMEOUT`	No	Timeout in seconds for AI calls (embed and LLM). LLM gets max of this and 60.	30
`MEMORY_AI_BASE_URL`	No	Base URL for the OpenAI-compatible API endpoint (e.g., Ollama, OpenAI, LM Studio).	http://localhost:11434/v1
`MEMORY_EMBED_MODEL`	No	Model name for generating embeddings.	nomic-embed-text
`MEMORY_LLM_BASE_URL`	No	Optional separate base URL for LLM calls (split backend). Defaults to MEMORY_AI_BASE_URL.
`MEMORY_DECAY_HALFLIFE_DAYS`	No	Global default half-life in days for memory confidence decay.	90
`MEMORY_TRUST_DEFAULT_REMEMBER`	No	Default trust tier for memories created via the remember tool (1-5).	5
`MEMORY_DECAY_CATEGORY_HALFLIFE`	No	JSON object mapping category names to custom half-life days (e.g., {"preference": 180}).
`MEMORY_LOCATION_DECAY_HALFLIFE_HOURS`	No	Half-life in hours for location memory confidence decay.	24

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

ActivityInactive

ResponsivenessNo issues