Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.murmur.dev/llms.txt

Use this file to discover all available pages before exploring further.

An environment is a catalog resource that defines the compute shape of agent VMs. Each workspace references one environment via its environment_ref field. Environments are substrate-aware but placement-agnostic — the same environment can be reused across any placement that shares the same substrate.

​
Fields

NameTypeRequiredDescription
namestringyesUnique identifier. DNS label format: [a-z][a-z0-9-]{0,62}.
substratestringyesCloud provider. Values: GCP, AWS, DOCKER.
machine_type_refstringyesName of a machine-type resource. Must exist and match this environment’s substrate.
disk_type_refstringconditionalName of a disk-type resource. Required for GCP and AWS substrates. Optional for DOCKER. Must exist and match this environment’s substrate.
disk_size_gbintnoBoot disk size in GiB. 0 uses the image default. Must be non-negative.
descriptionstringnoHuman-readable description shown in the dashboard. Maximum 1024 bytes.
The machine_type_ref and disk_type_ref must reference resources whose substrate matches the environment’s substrate. A substrate mismatch is rejected on write.

​
Examples

​
GCP environment

name: large
substrate: GCP
machine_type_ref: n2-standard-64
disk_type_ref: pd-ssd
disk_size_gb: 600
description: "64 vCPU, 256 GB RAM for large codebases"

cat <<'EOF' | murmur set environment large
name: large
substrate: GCP
machine_type_ref: n2-standard-64
disk_type_ref: pd-ssd
disk_size_gb: 600
description: "64 vCPU, 256 GB RAM for large codebases"
EOF

​
AWS environment

name: aws-default
substrate: AWS
machine_type_ref: m7i-xlarge
disk_type_ref: gp3
disk_size_gb: 200
description: "4 vCPU, 16 GB RAM general purpose"

​
Listing environments

murmur get environment

NAME          SUBSTRATE  MACHINE TYPE REF  DISK SIZE GB  DESCRIPTION
default       GCP        n2-standard-8     0             Platform default
large         GCP        n2-standard-64    600           64 vCPU, 256 GB RAM for large codebases
aws-default   AWS        m7i-xlarge        200           4 vCPU, 16 GB RAM general purpose

​
Reading a single environment

murmur get environment large

​
Errors

CodeMeaningWhat to do
INVALID_ARGUMENTname is requiredProvide a name field.
INVALID_ARGUMENTname must match [a-z][a-z0-9-]{0,62}Use a lowercase DNS label for the name.
INVALID_ARGUMENTsubstrate is requiredSet the substrate field to GCP, AWS, or DOCKER.
INVALID_ARGUMENTmachine_type_ref is requiredProvide a machine_type_ref that names an existing machine-type resource.
INVALID_ARGUMENTdisk_size_gb must be non-negativeSet disk_size_gb to 0 or a positive integer.
INVALID_ARGUMENTdisk_type_ref is required for GCP and AWS substratesProvide a disk_type_ref when the substrate is GCP or AWS.
INVALID_ARGUMENTdescription exceeds 1024 byte limitShorten the description to 1024 bytes or fewer.
INVALID_ARGUMENTmachine_type_ref "..." does not existThe referenced machine-type resource does not exist. Create it first or correct the name.
INVALID_ARGUMENTdisk_type_ref "..." does not existThe referenced disk-type resource does not exist. Create it first or correct the name.
INVALID_ARGUMENTsubstrate mismatch: machine type "..." is ... but environment substrate is ...The machine-type resource targets a different substrate than the environment. Use a machine type that matches.
INVALID_ARGUMENTsubstrate mismatch: disk type "..." is ... but environment substrate is ...The disk-type resource targets a different substrate than the environment. Use a disk type that matches.
FAILED_PRECONDITIONcannot delete environment "...": referenced by workspace: ...Remove the workspace reference before deleting the environment.