Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
SMOLVMNoPath to the `smolvm` binary. Used when the binary is not on PATH.
SMOL_MCP_CPUSNoNumber of CPUs allocated to each machine.2
SMOL_CLOUD_URLNoBase URL of the smol cloud REST API. Used when the `cloud` target is selected.
SMOL_LOCAL_URLNoListen address of an already-running `smolvm serve`. If set, the server uses this existing instance instead of starting its own.
SMOL_MCP_CONFIGNoPath to a JSON config file. Env vars take precedence over file values.
SMOL_CLOUD_TOKENNoBearer token for the smol cloud REST API. Required when using the `cloud` target.
SMOL_MCP_HTTP_HOSTNoBind host for the HTTP transport.127.0.0.1
SMOL_MCP_HTTP_PATHNoURL path for the HTTP transport./mcp
SMOL_MCP_HTTP_PORTNoPort for the HTTP transport.8080
SMOL_MCP_MEMORY_MBNoMemory allocated to each machine, in MiB.2048
SMOL_MCP_AUTH_TOKENNoAuthentication token required by the HTTP transport. The HTTP server refuses to start without it.
SMOL_MCP_MACHINE_PREFIXNoPrefix marking a machine as ephemeral.mcp-
SMOL_MCP_MAX_OUTPUT_BYTESNoMaximum output bytes kept per stream before truncation.65536
SMOL_MCP_RUN_ONCE_NETWORKNoNetwork mode for the `run-once` tool on the local target.open
SMOL_MCP_EXEC_TIMEOUT_SECSNoMaximum execution time for a command before it times out.120
SMOL_MCP_EPHEMERAL_TTL_SECSNoTime-to-live for ephemeral machines in seconds.3600
SMOL_MCP_READY_TIMEOUT_SECSNoTime to wait for a machine to become ready.120

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

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
list-machinesB

List machines on the target.

get-machineB

Get one machine's state and resources.

create-machineA

Create a machine from an OCI image, start it, and wait until commands run in it.

run-commandC

Run a command in a running machine. exitCode comes from the guest; a failing command is not an error.

run-onceA

Create a throwaway machine from an image, run one command, and delete the machine even on timeout.

read-fileB

Read a file from a machine.

write-fileA

Write a file into a machine. Waits until the workload container runs so the file is not lost.

stop-machineA

Stop a running machine (it can be started again with create-machine's name or the CLI).

delete-machineA

Delete a machine, running or not.

machine-logsB

Tail the machine's console log. Local only.

pull-imageC

Pull an image into a running machine's local cache. Local only.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources