FUSE 就是你所需的一切 - 通过文件系统赋予代理访问任何内容的权限
Giving agents access to a sandboxed environment with a shell and a filesystem has been the latest hype when it comes to agentic harnesses. Recent examples of this include:
给代理访问带有 shell 和文件系统的沙箱环境，这是代理框架方面的最新热潮。最近的例子包括：
- Turso’s AgentFS
- Turso’s AgentFS
- Anthropic’s Agent SDK, which brings Claude Code’s harness to non-coding domains
- Anthropic’s Agent SDK，它将 Claude Code 的 harness 带到非编码领域
- Vercel rebuilding their text-to-SQL agent on top of a sandbox
- Vercel 重建他们的 text-to-SQL 代理，基于 sandbox
- Anthropic’s Agent Skills for filesystem-based progressive disclosure
- Anthropic’s Agent Skills，用于基于文件系统的渐进式披露
The argument for why this is good goes something like this:
为什么这很好，其论点大致如下：
The big labs are doing heavy RL for coding tasks in these kinds of environments. Aligning more closely with such a harness brings free gains from the coding domain to other problem spaces.
大型实验室正在这些类型的环境中为编码任务进行大量 RL。更多地与这样的测试框架对齐，可以从编码领域免费获得收益并应用到其他问题空间。
Beyond that, replacing a bunch of search/write/move/list tools with a single Bash tool reduces the tool space significantly. Agents can chain operations together intuitively. Unix paradigms give you good tool design for free.
除此之外，用单个 Bash 工具替换大量的搜索/写入/移动/列出工具，大大减少了工具空间。代理可以直观地将操作串联在一起。Unix 范式免费为您提供良好的工具设计。
On top of that there is more nice patterns that emerge from having a filesystem, for example:
在此之上，还有更多从拥有文件系统而涌现的良好模式，例如：
- Plan/scratch files: Agents can create temporary files to organize their thoughts, track progress, or store intermediate results. This emerges naturally from having filesystem access, no need to design a separate “notepad” tool.
- Plan/scratch files：代理可以创建临时文件来组织他们的想法、跟踪进度或存储中间结果。这从拥有文件系统访问权限中自然出现，无需设计单独的 "notepad" 工具。
- Long context handling: As conversations grow, you can compact old messages and tool results into files on the filesystem. The agent can re-read them when needed rather than keeping everything in context.
- 长上下文处理：随着对话增长，您可以将旧消息和工具结果压缩到文件系统上的文件中。智能体可以在需要时重新读取它们，而不是将所有内容保留在上下文中。
So the advantages are clear. But how do you actually apply this to your domain? This is where t...