> ## 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.

# Core Concepts

> The key ideas behind Murmur — agents, workspaces, placements, images, and the catalog resources that connect them into a coding platform.

Murmur orchestrates autonomous AI coding agents on cloud infrastructure managed by the platform. You don't provision VMs or manage clusters — you sign in, connect your repos, and spawn agents.

## Organizations

Your **GitHub org** is your tenant. When an org admin installs the Macroscope GitHub App, it creates the tenant boundary. All resources — agents, workspaces, secrets, catalog entries — are namespaced to your org.

* **Personal tenants** exist automatically for every authenticated GitHub user
* **Organization tenants** exist when the Macroscope GitHub App is installed on a GitHub org

## Users

Users log in to Murmur with their GitHub account. They automatically have access to any organization that has installed the Macroscope App.

## Glossary

#### [Placements](/concepts/placement)

Where agents run — which cloud account, region, and network. Platform placements are preconfigured; customer placements let you run agents in your own cloud.

#### [Environments](/concepts/environments)

The VM compute shape — machine type, disk size. Each workspace references one environment.

#### [Recipes](/concepts/recipe)

A base image plus a provisioning script that installs your toolchain. Baking a recipe produces an image.

#### [Images](/concepts/images)

The result of baking a recipe. Agents boot from images, so everything your provisioning script installs is available immediately.

#### [Workspaces](/concepts/workspaces)

The top-level organizational unit within a tenant. Groups repositories, references an environment and placement, and defines defaults for agents spawned within it.

#### [Agents](/concepts/agents)

An autonomous Claude Code (or Codex) session running on a cloud VM. Agents clone repos, execute tasks, open pull requests, and can spawn child agents recursively.

#### [Agent Personas](/concepts/agent-personas)

Reusable templates that define how an agent behaves — prompt, model, tools, and constraints. The difference between spawning "a Claude" and spawning "an architect."

#### [Pools](/concepts/pools)

Pre-booted VMs that let agents start in seconds. The platform maintains warm VMs ready for instant assignment.

#### [Flights](/concepts/flights)

Markdown documents that define multi-agent workflows — task dependencies, personas, and event triggers. A pilot agent reads the flight and orchestrates the sub-agents.

#### [Events](/concepts/events)

Real-time GitHub events delivered to running agents — PR comments, reviews, CI results, and file changes. Agents react autonomously without polling.

#### [Secrets](/concepts/secrets)

Credentials and sensitive values — tenant-wide, per-developer, or per-spawn. Encrypted at rest, delivered as environment variables on agent VMs.

#### [Service Profiles](/concepts/service-profiles)

Non-human identities for automated agents — flights, scheduled tasks, and bots that operate on behalf of the org.

## More

#### [Security](/security/overview)

Authentication, authorization, and encryption — how Murmur protects credentials and enforces access control.

#### [Catalog resources](/catalog/overview)

The resource registry that stores all Murmur configuration. Every kind — workspaces, environments, placements, secrets, and more — documented field by field.

#### [CLI](/cli/overview)

Install and configure the `murmur` command-line interface. Every command documented with flags, examples, and errors.

#### [MCP server](/mcp-server/overview)

Use Murmur tools from within Claude Code sessions. Every MCP tool documented with parameters, responses, and examples.