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.

Launches an interactive onboarding session that creates a recipe, validates the provisioning script, and sets up a workspace for the target repo. Runs Claude Code with a specialized persona that guides you through each step.

​
Synopsis

murmur init [flags] [directory]

​
Arguments

NameTypeRequiredDescription
directorystringnoPath to the project root. Default: current directory.
--namestringnoRecipe name. Default: derived from the git remote (slugified repo name).
--workspace-namestringnoWorkspace name. Default: inherited from murmur.yaml, or falls back to the recipe name.
--environmentstringnoTarget environment. Default: discovered or prompted during the session.
--placementstringnoTarget placement. Default: discovered or prompted during the session.
--overwriteboolnoReplace an existing recipe and workspace with the same name without prompting. Default: false.

​
Prerequisites

murmur init checks these prerequisites before starting:
  • Project root — The directory contains at least one recognized project marker (.git, go.mod, package.json, pyproject.toml, Cargo.toml, Gemfile, or Makefile).
  • Claude Code — claude is on your PATH. Install with npm install -g @anthropic-ai/claude-code.
  • Claude authentication — claude auth status succeeds. Run claude auth login if not.
  • MCP server — The murmur MCP server is configured in Claude Code. Run claude mcp add murmur -- murmur mcp if not.
  • murmur setup — A valid .murmur/murmur.yaml exists. Run murmur setup first.

​
What it does

The onboarding session runs as an interactive Claude Code conversation. The persona drives a workflow that:
  1. Inspects the repo to understand languages, frameworks, and build tools.
  2. Creates a provisioning script (.murmur/provisioning.sh) that installs dependencies and prepares the environment.
  3. Validates the script by running a bake-iterate-fix loop until the image builds cleanly.
  4. Creates the recipe and workspace in the catalog.
  5. Optionally runs a smoke test by spawning a test agent.
After the session completes, murmur init persists the workspace name to murmur.yaml so that murmur spawn uses it by default.
The onboarding session runs on your machine, not in a sandbox. Read-only operations and writes to .murmur/ are auto-approved. Other operations prompt you for confirmation.

​
Examples

​
Onboard the current repo

murmur init

​
Onboard a different directory

murmur init /path/to/my-project

​
Onboard with explicit names

murmur init --name my-api --workspace-name backend

​
Replace an existing recipe and workspace

murmur init --overwrite

​
Errors

CodeMeaningWhat to do
no project detectedThe target directory has no recognized project marker.Run murmur init from a project root, or pass the project directory as an argument.
claude not foundClaude Code is not installed.Install with npm install -g @anthropic-ai/claude-code.
claude not authenticatedClaude Code is not logged in.Run claude auth login.
murmur MCP server not configuredThe murmur MCP server is not registered with Claude Code.Run claude mcp add murmur -- murmur mcp.
not configured — run: murmur setupNo .murmur/murmur.yaml found.Run murmur setup to create the config file.