> ## 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.

# 获取 Agent

> 通过 ID 或自定义 ID 检索 Agent

```
GET /openapi/memorylake/api/v3/agents/{id}
```

检索单个 Agent。默认情况下，`id` 路径参数与内部 Agent ID 匹配。设置 `by_custom_id=true` 可改为按自定义标识符查找 Agent。

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

### 路径参数

<ParamField path="id" type="string" required>
  Agent 标识符。默认作为内部 ID 解释，当 `by_custom_id` 为 `true` 时作为自定义 ID 解释。
</ParamField>

### 查询参数

<ParamField query="by_custom_id" type="boolean">
  当为 `true` 时，`id` 路径参数将与 Agent 的 `custom_id` 而非内部 `id` 进行匹配。
</ParamField>

<RequestExample>
  ```bash cURL theme={null}
  curl -X GET 'https://app.memorylake.ai/openapi/memorylake/api/v3/agents/agt_r8k2m1n3' \
    -H 'Authorization: Bearer sk_xxxxxx'
  ```
</RequestExample>

### 响应

<ResponseField name="data" type="object">
  <Expandable title="Agent 对象">
    <ResponseField name="id" type="string">Agent 唯一标识符</ResponseField>
    <ResponseField name="name" type="string">Agent 名称</ResponseField>
    <ResponseField name="description" type="string">Agent 描述</ResponseField>
    <ResponseField name="metadata" type="object">用于存储 Agent 附加信息的任意键值对</ResponseField>
    <ResponseField name="version" type="integer">当前活跃版本号</ResponseField>
    <ResponseField name="model" type="string">Agent 使用的模型</ResponseField>
    <ResponseField name="capabilities" type="array">Agent 支持的能力标识符列表</ResponseField>

    <ResponseField name="policies" type="object">
      约束 Agent 行为的执行策略

      <Expandable title="Policies 对象">
        <ResponseField name="max_turns" type="integer">每次运行的最大对话轮次数</ResponseField>
        <ResponseField name="max_tool_calls" type="integer">每轮的最大工具调用次数</ResponseField>
        <ResponseField name="allow_tools" type="array">Agent 可调用的工具白名单</ResponseField>
        <ResponseField name="deny_tools" type="array">Agent 不可调用的工具黑名单</ResponseField>
        <ResponseField name="subagent_timeout" type="integer">子 Agent 在被取消前可运行的最大秒数</ResponseField>
        <ResponseField name="subagent_max_concurrency" type="integer">可同时运行的最大子 Agent 数量</ResponseField>
      </Expandable>
    </ResponseField>

    <ResponseField name="output" type="object">
      Agent 的输出配置

      <Expandable title="Output 对象">
        <ResponseField name="mode" type="string">输出模式</ResponseField>
        <ResponseField name="json_schema" type="object">结构化输出的 JSON Schema</ResponseField>
      </Expandable>
    </ResponseField>

    <ResponseField name="subagents" type="array">
      附加到此 Agent 的子 Agent 定义

      <Expandable title="Subagent 对象">
        <ResponseField name="name" type="string">子 Agent 名称</ResponseField>
        <ResponseField name="mode" type="string">子 Agent 的执行模式</ResponseField>
        <ResponseField name="context" type="string">传递给子 Agent 的上下文指令</ResponseField>
        <ResponseField name="inherit_tools" type="boolean">子 Agent 是否继承父 Agent 的工具</ResponseField>
      </Expandable>
    </ResponseField>

    <ResponseField name="skills" type="array">
      附加到此 Agent 的技能

      <Expandable title="Skill 引用">
        <ResponseField name="skill_id" type="string">技能标识符</ResponseField>
        <ResponseField name="skill_version" type="integer">技能版本号</ResponseField>
      </Expandable>
    </ResponseField>

    <ResponseField name="custom_id" type="string">用于外部系统集成的自定义标识符</ResponseField>
    <ResponseField name="actor_id" type="string">与此 Agent 关联的 Actor 标识符</ResponseField>
    <ResponseField name="latest_version" type="integer">此 Agent 可用的最新版本号</ResponseField>
    <ResponseField name="system_prompt" type="string">用于指示 Agent 的系统提示词</ResponseField>
    <ResponseField name="model_settings" type="object">模型特定设置，如温度和最大 token 数</ResponseField>
    <ResponseField name="runtime_bindings" type="object">工具和数据源的运行时绑定配置</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>

<ResponseExample>
  ```json Success (200) theme={null}
  {
    "success": true,
    "data": {
      "id": "agt_r8k2m1n3",
      "name": "Support Assistant",
      "description": "Handles customer support inquiries using knowledge base",
      "metadata": {
        "team": "customer-success",
        "tier": "production"
      },
      "version": 3,
      "model": "gpt-4o",
      "capabilities": ["tool_use", "memory_read", "memory_write"],
      "policies": {
        "max_turns": 20,
        "max_tool_calls": 10,
        "allow_tools": ["search_kb", "create_ticket"],
        "deny_tools": [],
        "subagent_timeout": 300,
        "subagent_max_concurrency": 3
      },
      "output": {
        "mode": "text",
        "json_schema": null
      },
      "subagents": [
        {
          "name": "ticket-creator",
          "mode": "delegate",
          "context": "Create support tickets from conversation context",
          "inherit_tools": false
        }
      ],
      "skills": [
        {
          "skill_id": "skl_kb_search",
          "skill_version": 2
        }
      ],
      "custom_id": "support-assistant-v1",
      "actor_id": "act_p4q7w2x9",
      "latest_version": 3,
      "system_prompt": "You are a helpful support assistant. Use the knowledge base to answer questions accurately.",
      "model_settings": {
        "temperature": 0.7,
        "max_tokens": 4096
      },
      "runtime_bindings": {
        "knowledge_base": "kb_prod_docs"
      },
      "created_at": "2025-03-10T09:00:00Z",
      "updated_at": "2025-04-22T14:30:00Z"
    }
  }
  ```
</ResponseExample>