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

# search

> Search across your conversations and projects using natural language.

```bash theme={null}
memorybase search <query> [options]
```

Runs a semantic search over your entire MemoryBase. This is the same search that powers the browser extension's context injection -- it finds the most relevant conversation snippets and project layers based on meaning, not just keywords.

## Arguments

| Argument | Required | Description                                    |
| -------- | -------- | ---------------------------------------------- |
| `query`  | Yes      | What you want to search for (natural language) |

## Options

| Option             | Description                                              |
| ------------------ | -------------------------------------------------------- |
| `--top-layers <n>` | Maximum number of matched layers to return (1-10)        |
| `--pinned <ids>`   | Comma-separated project IDs to always include in results |
| `--format <fmt>`   | Output format: `table` (default), `json`, or `markdown`  |

## Examples

**Basic search:**

```bash theme={null}
memorybase search "react authentication flow"
```

**Get results as JSON (useful for piping to other tools):**

```bash theme={null}
memorybase search "database schema" --format json
```

**Limit to top 3 results:**

```bash theme={null}
memorybase search "API error handling" --top-layers 3
```

**Pin a specific project to always appear in results:**

```bash theme={null}
memorybase search "deployment" --pinned proj-abc123,proj-def456
```