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

# 搜索记忆

> 使用自然语言在工作空间中跨所有记忆源进行搜索

```
POST /openapi/memorylake/api/v3/workspaces/{workspaceId}/memories/search
```

使用自然语言查询在工作空间中跨所有记忆源（文档和事实）进行搜索。结果按来源类型分组，便于你在 UI 中分别展示各类别或将其合并。使用可选筛选条件可将结果限定到特定项目、Actor 或记忆类型。

<Note>
  **所需权限：** `workspace:read`
</Note>

### 路径参数

<ParamField path="workspaceId" type="string" required>
  工作空间标识符
</ParamField>

### 请求体

<ParamField body="query" type="string" required>
  自然语言搜索查询
</ParamField>

<ParamField body="project_ids" type="array of strings">
  将搜索限定到特定项目。省略时将搜索工作空间中的所有项目。
</ParamField>

<ParamField body="actor_ids" type="array of strings">
  将搜索限定到与特定 Actor 关联的记忆
</ParamField>

<ParamField body="memory_types" type="array of strings">
  按记忆类型筛选结果。可接受的值：`document`、`fact`。省略时将搜索所有类型。
</ParamField>

<ParamField body="top_k" type="integer">
  每个来源类型返回的最大结果数量
</ParamField>

<RequestExample>
  ```bash cURL theme={null}
  curl -X POST 'https://app.memorylake.ai/openapi/memorylake/api/v3/workspaces/ws_abc123/memories/search' \
    -H 'Authorization: Bearer sk_xxxxxx' \
    -H 'Content-Type: application/json' \
    -d '{
      "query": "What were the quarterly revenue figures?",
      "project_ids": ["proj_def456"],
      "memory_types": ["document", "fact"],
      "top_k": 10
    }'
  ```
</RequestExample>

### 响应

<ResponseField name="data" type="object">
  <Expandable title="按来源类型分组的搜索结果">
    <ResponseField name="documents" type="array">
      文档搜索结果，按源文档分组

      <Expandable title="DocumentSearchResult">
        <ResponseField name="document_id" type="string">文档标识符</ResponseField>
        <ResponseField name="document_name" type="string">文档的显示名称</ResponseField>
        <ResponseField name="file_name" type="string">原始文件名</ResponseField>
        <ResponseField name="document_summary" type="string">文档摘要</ResponseField>
        <ResponseField name="source_type" type="string">文档的来源类型（例如 `upload`、`connector`）</ResponseField>
        <ResponseField name="sheet_name" type="string">工作表名称（当文档为工作簿时）</ResponseField>

        <ResponseField name="items" type="array">
          文档中匹配的内容项
        </ResponseField>
      </Expandable>
    </ResponseField>

    <ResponseField name="facts" type="array">
      事实搜索结果

      <Expandable title="FactSearchResult">
        <ResponseField name="id" type="string">事实标识符</ResponseField>
        <ResponseField name="fact" type="string">事实文本</ResponseField>
        <ResponseField name="score" type="number">相关性分数</ResponseField>
        <ResponseField name="metadata" type="object">与事实关联的附加元数据</ResponseField>
        <ResponseField name="created_at" type="string">创建时间戳</ResponseField>
        <ResponseField name="updated_at" type="string">最后更新时间戳</ResponseField>
      </Expandable>
    </ResponseField>
  </Expandable>
</ResponseField>

<ResponseExample>
  ```json Success (200) theme={null}
  {
    "success": true,
    "data": {
      "documents": [
        {
          "document_id": "doc-abc123",
          "document_name": "Q1 2024 Financial Report",
          "file_name": "q1_2024_report.pdf",
          "document_summary": "Quarterly financial report covering revenue, expenses, and projections for Q1 2024.",
          "source_type": "upload",
          "sheet_name": null,
          "items": [
            {
              "text": "Q1 2024 revenue reached $12.5M, a 15% increase over Q4 2023.",
              "range": "p8"
            }
          ]
        }
      ],
      "facts": [
        {
          "id": "fact-xyz789",
          "fact": "Company quarterly revenue for Q1 2024 was $12.5M",
          "score": 0.92,
          "metadata": {},
          "created_at": "2024-04-01T09:00:00Z",
          "updated_at": "2024-04-01T09:00:00Z"
        }
      ]
    }
  }
  ```
</ResponseExample>