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.

Murmur setup happens in two stages: an admin initially configures the team workspace once [see admin quickstart guide], then each developer sets up Murmur locally [see below].
An admin must configure the workspace before developers can use Murmur.

​
Developer setup

Complete the steps below after your admin has finished workspace setup [see admin quickstart guide]. CLI setup requires a GitHub account, gh CLI, and a Claude API key or Anthropic subscription.

​
Step 1: Install the CLI

brew tap prassoai/tap
brew install prassoai/tap/murmur

​
Step 2: Setup your credentials

murmur setup
This sets up your credentials and bookmarks your tenant in .murmur/murmur.yaml .

​
Step 3: Add the Murmur MCP

claude mcp add murmur murmur mcp
This binds your CLI/MCP to your organization.

​
Step 4: Spawn your first agent

There are two ways to spawn an agent from the command line: either by using the Murmur CLI or by asking your coding agent.

​
Option 1: Spawn via Murmur CLI

murmur spawn my-first-task "Create a file called hello.txt with 'Hello from Murmur'" --out pr

​
Option 2: Spawn from your coding agent (Claude)

murmur director

​
Next steps

Core Concepts

Understand the architecture, agent lifecycle, and tenant model.

Flights

Orchestrate multiple agents with declarative Markdown DAGs.

CLI Reference

All flags and options for murmur spawn.

MCP Server

How agents use MCP tools to spawn children and coordinate work.