> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentscope.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Overview

> Choose how an application or agent should run and use ReMe.

Every usage mode works with the same ReMe workspace and jobs. The main decisions are who owns the ReMe lifecycle and whether the host needs a shared service, a service-based agent integration, or in-process Python access.

## Choose a Path

Use the integration boundary that best matches the host:

| Path                                             | Choose it when                                              | Responsibility                                   |
| ------------------------------------------------ | ----------------------------------------------------------- | ------------------------------------------------ |
| [Services](/reme/latest/en/integration/services) | ReMe should run independently and serve one or more clients | Expose MCP, CLI, or HTTP access                  |
| [Python](/reme/latest/en/integration/python)     | A Python or AgentScope application needs in-process ReMe    | Call `ReMe.run_job()` or use `ReMeMiddleware`    |
| [Agents](/reme/latest/en/integration/agents)     | An agent should connect to a deployed ReMe service          | Use tools, a plugin, or a Skill over the service |

Choose [Python](/reme/latest/en/integration/python) for AgentScope's first-party `ReMeMiddleware` or when another Python host should own ReMe in process. Choose [Agents](/reme/latest/en/integration/agents) for service-based agent integrations, and review [Services](/reme/latest/en/integration/services) when selecting MCP, CLI, or HTTP.

## Runtime Model

Direct Python and AgentScope embed ReMe in the host process. MCP, CLI, and HTTP access a separately running ReMe service. A Skill sits above those service interfaces and tells an agent when to use them.

```text theme={null}
Python host ── ReMe.run_job() ─┐
AgentScope ── ReMeMiddleware ──┴──> ReMe Application ──> workspace

Agent / app ── MCP / CLI / HTTP ──> ReMe service ─────> workspace
                 ↑
              Skill rules
```

Keep one runtime owner for a workspace: either let the Python host manage the application lifecycle, or start one shared service with `reme start`. The runtime owns background indexing and scheduled consolidation; the host still owns session hooks, stable session identifiers, and decisions about when to recall or record.