Home 零对冲(ZeroHedge)每日HackerNews

ChatGPT容器现在可以运行bash，安装pip/npm包，以及下载文件。
ChatGPT Containers can now run bash, pip/npm install packages and download files

原始链接: https://simonwillison.net/2026/Jan/26/chatgpt-containers/

## ChatGPT 代码解释器获得重大升级 ChatGPT 的“高级数据分析”功能（前身为代码解释器）获得了一次重要的、未公开的升级，极大地扩展了其编码和数据处理能力。现在，除了 Python 之外，ChatGPT 还可以直接执行 **Bash、Node.js (JavaScript) 甚至 Ruby、Perl 和 Java** 等语言的代码。 一个关键的补充是 **`container.download` 工具**，它允许 ChatGPT 从公开可访问的 URL 获取文件并在其沙盒环境中保存。虽然容器仍然缺乏直接的互联网访问权限，但现在可以通过自定义代理使用 **`pip` 和 `npm`** 安装软件包。 这次升级是通过实验发现的，揭示了启用软件包安装的代理系统 (`applied-caas-gateway1.internal.api.openai.org`) 的细节。安全测试表明，`container.download` 工具目前可以防止提示注入攻击，要求 URL 要么由用户直接输入，要么来自安全搜索结果。 这些增强功能使 ChatGPT 能够分析来自网络的数据，利用更广泛的编程语言，并利用外部软件包——使其成为一个更加通用的编码和数据分析工具。然而，作者敦促 OpenAI 为这些新功能提供官方文档，目前这些功能被称为“ChatGPT 容器”。

## ChatGPT 更新与开发者讨论 ChatGPT 的最新更新允许容器运行 bash，安装软件包 (pip/npm)，并下载文件，引发了关于软件开发未来的讨论。一个关键主题是向“本地”工具调用的转变，可能绕过沙箱限制并实现更强大的功能。 评论员推测 ChatGPT 生成的单次使用、按需应用程序将会兴起，以及像 Go 这样的编译型语言可能会回归。 观点是，随着 LLM 成为主要的编码者，动态语言（Python、JavaScript）的优势会减弱，因为 LLM 可以在拥有足够训练数据的情况下处理编译型语言。 目前这种转变是否现实存在争议，但可移植二进制文件的便利性很有吸引力。 同时也提出了对易于携带的恶意软件的担忧。 值得注意的是，标准 ChatGPT 现在也支持在包括 Java、Ruby 等多种语言中执行代码，尽管这些更新尚未正式公布。
相关文章
原文

26th January 2026

One of my favourite features of ChatGPT is its ability to write and execute code in a container. This feature launched as ChatGPT Code Interpreter nearly three years ago, was half-heartedly rebranded to “Advanced Data Analysis” at some point and is generally really difficult to find detailed documentation about. Case in point: it appears to have had a massive upgrade at some point in the past few months, and I can’t find documentation about the new capabilities anywhere!

Here are the most notable new features:

  1. ChatGPT can directly run Bash commands now. Previously it was limited to Python code only, although it could run shell commands via the Python subprocess module.
  2. It has Node.js and can run JavaScript directly in addition to Python. I also got it to run “hello world” in Ruby, Perl, PHP, Go, Java, Swift, Kotlin, C and C++. No Rust yet though!
  3. While the container still can’t make outbound network requests, pip install package and npm install package both work now via a custom proxy mechanism.
  4. ChatGPT can locate the URL for a file on the web and use a container.download tool to download that file and save it to a path within the sandboxed container.

This is a substantial upgrade! ChatGPT can now write and then test code in 10 new languages (11 if you count Bash), can find files online and download them into the container, and can install additional packages via pip and npm to help it solve problems.

(OpenAI really need to develop better habits at keeping their release notes up-to-date!)

I was initially suspicious that maybe I’d stumbled into a new preview feature that wasn’t available to everyone, but I tried some experiments in a free ChatGPT account and confirmed that the new features are available there as well.

container.download

My first clue to the new features came the other day when I got curious about Los Angeles air quality, in particular has the growing number of electric vehicles there hade a measurable impact?

I prompted a fresh GPT-5.2 Thinking session with:

Show me Los Angeles air quality over time for last 20 years

Here’s the shared transcript. Annoyingly that doesn’t make the thinking trace available, so here’s a screenshot:

Exploring methods to fetch AQI data and reports Downloading AQI data for Los Angeles I found an xlsx file titled "Air Quality Statistics by City, 2000-2023", which should include data for Los Angeles. While user requested data for Using provided URL to download file The URL from the error message leads to an Excel file containing air quality trends. I can use the container.download tool to fetch it. Let’s proceed with this approach and see if it works out for downloading the file directly! Analyzed Parsing and analyzing the data for Los Angeles Now that we've downloaded the data, I'll parse it using Python. First, I'll focus on identifying the row(s) corresponding to Los Angeles, which may be labeled under "Los Angeles-Long Beach-Anaheim, CA." Afterward, I'll plot metrics like "Median AQI" and "Days Good" over time to visualize trends. Let’s begin!

This bit instantly caught my eye:

[...] The URL from the error message leads to an Excel file containing air quality trends. I can use the container.download tool to fetch it. [...]

What’s container.download? I asked ChatGPT:

Without running any searches tell me about your container.download tool

And got back:

Sure. container.download is a built-in tool I can call that fetches a file from a URL and saves it into the sandboxed container filesystem for this chat session.

What it does

  • Takes a publicly reachable URL and a destination filepath in the container.
  • Downloads the bytes from that URL and writes them to the given path.
  • After that, I can read/process the file locally in the container (e.g., unzip it, parse it with Python, open it as an image, convert it, etc.).

Signature (inputs)

It’s called like this (conceptually):

{
  "url": "https://example.com/somefile.zip",
  "filepath": "/mnt/data/somefile.zip"
}

Further experimentation confirmed that container.download behaves as described.

I had it download one file from a server with logs enabled and captured the IP address and user-agent. Here are the most interesting headers:

User-Agent: Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; ChatGPT-User/1.0; +https://openai.com/bot
Accept: text/html, application/xhtml+xml, application/xml;q=0.9, image/avif, image/webp, image/apng, */*;q=0.8, application/signed-exchange;v=b3;q=0.9
Cf-Connecting-Ip: 52.230.164.178

That 52.230.164.178 IP address resolves to Microsoft Azure Cloud (centralus) in Des Moines, Iowa.

Is container.download a data exfiltration vulnerability?

On the one hand, this is really useful! ChatGPT can navigate around websites looking for useful files, download those files to a container and then process them using Python or other languages.

Is this a data exfiltration vulnerability though? Could a prompt injection attack trick ChatGPT into leaking private data out to a container.download call to a URL with a query string that includes sensitive information?

I don’t think it can. I tried getting it to assemble a URL with a query string and access it using container.download and it couldn’t do it. It told me that it got back this error:

ERROR: download failed because url not viewed in conversation before. open the file or url using web.run first.

This looks to me like the same safety trick used by Claude’s Web Fetch tool: only allow URL access if that URL was either directly entered by the user or if it came from search results that could not have been influenced by a prompt injection.

(I poked at this a bit more and managed to get a simple constructed query string to pass through web.run—a different tool entirely—but when I tried to compose a longer query string containing the previous prompt history a web.run filter blocked it.)

So I think this is all safe, though I’m curious if it could hold firm against a more aggressive round of attacks from a seasoned security researcher.

Bash and other languages

The key lesson from coding agents like Claude Code and Codex CLI is that Bash rules everything: if an agent can run Bash commands in an environment it can do almost anything that can be achieved by typing commands into a computer.

When Anthropic added their own code interpreter feature to Claude last September they built that around Bash rather than just Python. It looks to me like OpenAI have now done the same thing for ChatGPT.

Here’s what ChatGPT looks like when it runs a Bash command—here my prompt was:

npm install a fun package and demonstrate using it

Screenshot of ChatGPT 5.2 Thinking interface with Activity panel. Main chat shows user prompt "npm install a fun package and demonstrate using it" and response "Thought for 32s > Yep — here's a fun one: cowsay 🐄 I installed it with npm and ran a tiny Node script:" followed by bash code block containing "mkdir -p /mnt/data/npmfun cd /mnt/data/npmfun npm init -y npm install cowsay --no-fund --no-audit node -e "const cowsay=require('cowsay'); console.log(cowsay.say({text:'Hel" and Output: section showing ASCII cow art saying "< Hello from npm inside the container! >" with cow drawing made of characters. Activity panel on right (32s) shows: "Running npm install in container" with bash command and output "v22.16.0 10.9.2", "Inspecting npm command output and container status" with directory listing showing "drwxr-xr-x 2 root oai_shared 80 Jan 26 17:32 . drwxrwxr-x 2 oai oai_shared 60 Jan 26 17:32 .. -rw-r--r-- 1 root root", "Re-running npm install" note about node_modules missing, and "Rerunning npm install with output" showing "added 41 packages in 2s".

It’s useful to click on the “Thinking” or “Thought for 32s” links as that opens the Activity sidebar with a detailed trace of what ChatGPT did to arrive at its answer. This helps guard against cheating—ChatGPT might claim to have run Bash in the main window but it can’t fake those black and white logs in the Activity panel.

I had it run Hello World in various languages later in that same session.

Terminal screenshot showing "Hello, world" programs in multiple languages. Top section shows code: swift -e 'print("Hello, world from Swift!")' followed by cat > hello.kts <<'KTS' / println("Hello, world from Kotlin!") / KTS / kotlinc -script hello.kts, then cat > hello.c <<'C' / #include <stdio.h> / int main(){ puts("Hello, world from C!"); return 0; } / C / gcc hello.c -o hello_c / ./hello_c, then cat > hello.cpp <<'CPP' / #include <iostream> / int main(){ std::cout << "Hello, world from C++!" << std::endl; } / CPP / g++ hello.cpp -o hello_cpp / ./hello_cpp. Bottom section shows output: Hello, world from Ruby! / Hello, world from Perl! / Hello, world from PHP! / Hello, world from Go! / Hello, world from Java! / Hello, world from Swift! / Hello, world from Kotlin! / Hello, world from C! / Hello, world from C++!. UI shows "Thought for 2m 29s" and "Done" at bottom.

Installing packages from pip and npm

In the previous example ChatGPT installed the cowsay package from npm and used it to draw an ASCII-art cow. But how could it do that if the container can’t make outbound network requests?

In another session I challenged it to explore its environment. and figure out how that worked.

Here’s the resulting Markdown report it created.

The key magic appears to be a applied-caas-gateway1.internal.api.openai.org proxy, available within the container and with various packaging tools configured to use it.

The following environment variables cause pip and uv to install packages from that proxy instead of directly from PyPI:

PIP_INDEX_URL=https://reader:****@packages.applied-caas-gateway1.internal.api.openai.org/.../pypi-public/simple
PIP_TRUSTED_HOST=packages.applied-caas-gateway1.internal.api.openai.org
UV_INDEX_URL=https://reader:****@packages.applied-caas-gateway1.internal.api.openai.org/.../pypi-public/simple
UV_INSECURE_HOST=https://packages.applied-caas-gateway1.internal.api.openai.org

This one appears to get npm to work:

NPM_CONFIG_REGISTRY=https://reader:****@packages.applied-caas-gateway1.internal.api.openai.org/.../npm-public

And it reported these suspicious looking variables as well:

CAAS_ARTIFACTORY_BASE_URL=packages.applied-caas-gateway1.internal.api.openai.org
CAAS_ARTIFACTORY_PYPI_REGISTRY=.../artifactory/api/pypi/pypi-public
CAAS_ARTIFACTORY_NPM_REGISTRY=.../artifactory/api/npm/npm-public
CAAS_ARTIFACTORY_GO_REGISTRY=.../artifactory/api/go/golang-main
CAAS_ARTIFACTORY_MAVEN_REGISTRY=.../artifactory/maven-public
CAAS_ARTIFACTORY_GRADLE_REGISTRY=.../artifactory/gradle-public
CAAS_ARTIFACTORY_CARGO_REGISTRY=.../artifactory/api/cargo/cargo-public/index
CAAS_ARTIFACTORY_DOCKER_REGISTRY=.../dockerhub-public
CAAS_ARTIFACTORY_READER_USERNAME=reader
CAAS_ARTIFACTORY_READER_PASSWORD=****
NETWORK=caas_packages_only

Neither Rust nor Docker are installed in the container environment, but maybe those registry references are a clue of features still to come.

Adding it all together

The result of all of this? You can tell ChatGPT to use Python or Node.js packages as part of a conversation and it will be able to install them and apply them against files you upload or that it downloads from the public web. That’s really cool.

The big missing feature here should be the easiest to provide: we need official documentation! A release notes entry would be a good start, but there are a lot of subtle details to how this new stuff works, its limitations and what it can be used for.

As always, I’d also encourage them to come up with a name for this set of features that properly represents how it works and what it can do.

In the meantime, I’m going to call this ChatGPT Containers.

联系我们 contact @ memedata.com