权限

使用声明式权限规则控制深度智能体的文件系统访问

使用声明式权限规则控制智能体可以读写哪些文件和目录。将规则列表传入 permissions=，智能体的内置文件系统工具就会遵守这些规则。

权限需要 deepagents>=1.9.1。

权限仅适用于内置文件系统工具（ls、read_file、glob、grep、write_file、edit_file）。自定义工具和访问文件系统的 MCP 工具不受覆盖。权限也不适用于沙箱后端，因为沙箱通过 execute 工具支持任意命令执行。

当你需要对内置文件系统工具的基于路径的允许/拒绝规则时，请使用 permissions。当你需要自定义验证逻辑（速率限制、审计日志、内容检查）或需要控制自定义工具时，请使用后端策略钩子。

基本用法

将 FilesystemPermission 规则列表传入 createDeepAgent。规则按声明顺序评估。第一条匹配的规则生效。如果没有规则匹配，则允许操作。

const agent = createDeepAgent({
  model,
  backend,
  permissions: [
    {
      operations: ["write"],
      paths: ["/**"],
      mode: "deny",
    },
  ],
});
if (!agent) throw new Error("basic: agent not created");

规则结构

每个 FilesystemPermission 有三个字段：

字段	类型	描述
operations	("read" | "write")[]	此规则适用的操作。"read" 覆盖 ls、read_file、glob、grep。"write" 覆盖 write_file、edit_file。
paths	string[]	用于匹配文件路径的 glob 模式（例如 ["/workspace/**"]）。支持 ** 递归匹配和 {a,b} 交替。
mode	"allow" | "deny"	是允许还是拒绝匹配的操作。默认为 "allow"。

规则使用先匹配优先评估：第一条 operations 和 paths 匹配当前调用的规则决定结果。如果没有规则匹配，调用将被允许（宽容默认值）。

路径必须是绝对路径（以 / 开头），不能包含 .. 或 ~。无效路径会在智能体构建时抛出异常。

示例

隔离到工作区目录

仅允许在 /workspace/ 下进行读写，拒绝其他所有操作：

const agent = createDeepAgent({
  model,
  backend,
  permissions: [
    {
      operations: ["read", "write"],
      paths: ["/workspace/**"],
      mode: "allow",
    },
    {
      operations: ["read", "write"],
      paths: ["/**"],
      mode: "deny",
    },
  ],
});
if (!agent) throw new Error("isolate-workspace: agent not created");

保护特定文件

const agent = createDeepAgent({
  model,
  backend,
  permissions: [
    {
      operations: ["read", "write"],
      paths: ["/workspace/.env", "/workspace/examples/**"],
      mode: "deny",
    },
    {
      operations: ["read", "write"],
      paths: ["/workspace/**"],
      mode: "allow",
    },
    {
      operations: ["read", "write"],
      paths: ["/**"],
      mode: "deny",
    },
  ],
});
if (!agent) throw new Error("protect-files: agent not created");

只读记忆

允许智能体读取记忆文件但阻止其修改。这对于组织范围的策略或共享知识库很有用，这些内容只应由应用代码更新。更多上下文请参阅只读与可写记忆。

const store = new InMemoryStore();
const agent = createDeepAgent({
  model,
  backend: new CompositeBackend(new StateBackend(), {
    "/memories/": new StoreBackend({
      namespace: (rt) => [rt.serverInfo.user.identity],
    }),
    "/policies/": new StoreBackend({
      namespace: (rt) => [rt.context.orgId],
    }),
  }),
  permissions: [
    {
      operations: ["write"],
      paths: ["/memories/**", "/policies/**"],
      mode: "deny",
    },
  ],
  store,
});
if (!agent) throw new Error("read-only-memory: agent not created");

拒绝所有访问

阻止所有读写。这是一个限制性基线，你可以在其之上叠加更具体的允许规则：

const agent = createDeepAgent({
  model,
  backend,
  permissions: [
    {
      operations: ["read", "write"],
      paths: ["/**"],
      mode: "deny",
    },
  ],
});
if (!agent) throw new Error("deny-all: agent not created");

规则排序

由于先匹配优先，规则顺序很重要。将更具体的规则放在更宽泛的规则之前：

const correctPermissions: FilesystemPermission[] = [
  { operations: ["read", "write"], paths: ["/workspace/.env"], mode: "deny" },
  {
    operations: ["read", "write"],
    paths: ["/workspace/**"],
    mode: "allow",
  },
  { operations: ["read", "write"], paths: ["/**"], mode: "deny" },
];

const incorrectPermissions: FilesystemPermission[] = [
  {
    operations: ["read", "write"],
    paths: ["/workspace/**"],
    mode: "allow",
  },
  {
    operations: ["read", "write"],
    paths: ["/workspace/.env"],
    mode: "deny",
  },
  { operations: ["read", "write"], paths: ["/**"], mode: "deny" },
];

子智能体权限

子智能体默认继承父智能体的权限。要给子智能体不同的权限，请在其规格中设置 permissions 字段。这将完全替换父级的规则。

const agent = createDeepAgent({
  model,
  backend,
  permissions: [
    {
      operations: ["read", "write"],
      paths: ["/workspace/**"],
      mode: "allow",
    },
    { operations: ["read", "write"], paths: ["/**"], mode: "deny" },
  ],
  subagents: [
    {
      name: "auditor",
      description: "Read-only code reviewer",
      systemPrompt: "Review the code for issues.",
      permissions: [
        { operations: ["write"], paths: ["/**"], mode: "deny" },
        { operations: ["read"], paths: ["/workspace/**"], mode: "allow" },
        { operations: ["read"], paths: ["/**"], mode: "deny" },
      ],
    },
  ],
});
if (!agent) throw new Error("subagent: agent not created");

要显式授予子智能体不受限制的访问，请设置 permissions: []。空数组会用无限制覆盖父级规则。省略 permissions 则从父级继承。

复合后端

当使用带沙箱默认后端的 CompositeBackend 时，每个权限路径必须限定在已知的路由前缀下。沙箱支持任意命令执行，因此仅基于路径的限制无法阻止通过 shell 命令的文件系统访问。将权限限定到特定路由的后端可以避免这种冲突。

const sandbox = new StateBackend();
const memoriesBackend = new StateBackend();
const composite = new CompositeBackend(sandbox, {
  "/memories/": memoriesBackend,
});
const agent = createDeepAgent({
  model,
  backend: composite,
  permissions: [
    { operations: ["write"], paths: ["/memories/**"], mode: "deny" },
  ],
});
if (!agent) throw new Error("composite-backend: agent not created");

包含任何路由之外路径的权限会在构建时抛出异常：

const sandbox = new StateBackend();
const memoriesBackend = new StateBackend();
const composite = new CompositeBackend(sandbox, {
  "/memories/": memoriesBackend,
});

createDeepAgent({
  model,
  backend: composite,
  permissions: [
    { operations: ["write"], paths: ["/workspace/**"], mode: "deny" },
  ],
});

createDeepAgent({
  model,
  backend: composite,
  permissions: [{ operations: ["read"], paths: ["/**"], mode: "deny" }],
});

---

连接这些文档到 Claude、VSCode 等工具，通过 MCP 获取实时解答。

在 GitHub 上编辑此页面或提交问题。