Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
MCP_PORTNoHost-side published port for the MCP service. Container always listens on 8000. Default host port is 8102 when unset.8102
SERVER_IPNoNAS IP to add to Host validation. Set together with MCP_BIND_IP for LAN access.
MCP_BIND_IPNoHost interface to bind the MCP service to. Default 127.0.0.1; set to the NAS LAN IP for Windows/LAN access.127.0.0.1
OPENAI_MODELNoModel used for image analysis.gpt-4.1-mini
OPENAI_API_KEYYesOpenAI API key. Required; the server will not start if empty.
VISION_USAGE_DBNoPath to the SQLite usage database for persistent call-count limits. Used when running server.py directly for local development.
VISION_IMAGE_DIRNoNAS path to the image-only shared folder, used as the bind mount source (e.g. /volume1/vision-images).
VISION_IMAGE_GIDNoOptional numeric GID of the shared folder owning group, added as a supplementary group for read access to images.
VISION_MCP_TOKENYesRandom string of at least 32 characters used as the Bearer token for HTTP MCP authentication. Required; the server will not start if empty. Must match the Authorization header value configured in Strata.
MCP_ALLOWED_HOSTSNoComma-separated allowed Host header values. Example: localhost:*,127.0.0.1:*
VISION_IMAGE_ROOTNoContainer/local path where images are read from. Used when running server.py directly for local development.
VISION_MAX_CALLS_PER_DAYNoMaximum API calls per UTC day. Default 20.20
VISION_MAX_OUTPUT_TOKENSNoMaximum output tokens. Default 800; allowed range 128-2000.800
VISION_MAX_CALLS_PER_10_MINNoMaximum API calls per rolling 10 minutes. Default 3.3

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

NameDescription

No tools

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources