This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

This article will describe how to download an image from a (docker) container registry.

Manual Download of Container Images with wget and curl

Intro

Remember the good `'ol days when you could just download software by visiting a website and click "download"?

Even apt and yum repositories were just simple HTTP servers that you could just curl (or wget) from. Using the package manager was, of course, more secure and convenient -- but you could always just download packages manually, if you wanted.

But have you ever tried to curl an image from a container registry, such as docker? Well friends, I have tried. And I have the scars to prove it.

It was a remarkably complex process that took me weeks to figure-out. Lucky you, this article will break it down.

Examples

Specifically, we'll look at how to download files from two OCI registries.

1. Docker Hub
2. GitHub Packages

Terms

First, here's some terminology used by OCI

1. OCI - Open Container Initiative
2. blob - A "blob" in the OCI spec just means a file
3. manifest - A "manifest" in the OCI spec means a list of files

Prerequisites

This guide was written in 2024, and it uses the following software and versions:

1. debian 12 (bookworm)
2. curl 7.88.1
3. OCI Distribution Spec v1.1.0 (which, unintuitively, uses the '/v2/' endpoint)

Of course, you'll need 'curl' installed. And, to parse json, 'jq' too.

sudo apt-get install curl jq

What is OCI?

OCI stands for Open Container Initiative.

OCI was originally formed in June 2015 for Docker and CoreOS. Today it's a wider, general-purpose (and annoyingly complex) way that many projects host files (that are extremely non-trivial to download).

One does not simply download a file from an OCI-complianet container registry. You must:

1. Generate an authentication token for the API
2. Make an API call to the registry, requesting to download a JSON "Manifest"
3. Parse the JSON Manifest to figure out the hash of the file that you want
4. Determine the download URL from the hash
5. Download the file (which might actually be many distinct file "layers")

One does not simply download from a container registry

In order to figure out how to make an API call to the registry, you must first read (and understand) the OCI specs here.

• https://opencontainers.org/release-notices/overview/

OCI APIs

OCI maintains three distinct specifications:

1. image spec
2. runtime spec
3. distribution spec

OCI "Distribution Spec" API

To figure out how to download a file from a container registry, we're interested in the "distribution spec". At the time of writing, the latest "distribution spec" can be downloaded here:

• https://github.com/opencontainers/distribution-spec/releases/tag/v1.1.0
• https://github.com/opencontainers/distribution-spec/releases/download/v1.1.0/oci-distribution-spec-v1.1.0.pdf

The above PDF file defines a set of API endpoints that we can use to query, parse, and then figure out how to download a file from a container registry. The table from the above PDF is copied below:

ID	Method	API Endpoint	Success	Failure
end-1	GET	/v2/	200	404/401
end-2	GET / HEAD	/v2/<name>/blobs/<digest>	200	404
end-3	GET / HEAD	/v2/<name>/manifests/<reference>	200	404
end-4a	POST	/v2/<name>/blobs/uploads/	202	404
end-4b	POST	/v2/<name>/blobs/uploads/?digest=<digest>	201/202	404/400
end-5	PATCH	/v2/<name>/blobs/uploads/<reference>	202	404/416
end-6	PUT	/v2/<name>/blobs/uploads/<reference>?digest=<digest>	201	404/400
end-7	PUT	/v2/<name>/manifests/<reference>	201	404
end-8a	GET	/v2/<name>/tags/list	200	404
end-8b	GET	/v2/<name>/tags/list?n=<integer>&last=<integer>	200	404
end-9	DELETE	/v2/<name>/manifests/<reference>	202	404/400/405
end-10	DELETE	/v2/<name>/blobs/<digest>	202	404/405
end-11	POST	/v2/<name>/blobs/uploads/?mount=<digest>&from=<other_name>	201	404
end-12a	GET	/v2/<name>/referrers/<digest>	200	404/400
end-12b	GET	/v2/<name>/referrers/<digest>?artifactType=<artifactType>	200	404/400
end-13	GET	/v2/<name>/blobs/uploads/<reference>	204	404

In OCI, files are (cryptically) called "blobs". In order to figure out the file that we want to download, we must first reference the list of files (called a "manifest").

The above table shows us how we can download a list of files (manifest) and then download the actual file (blob).

Examples

Let's look at how to download files from a couple different OCI registries:

1. Docker Hub
2. GitHub Packages

Docker Hub

To see the full example of downloading images from docker hub, click here

GitHub Packages

To see the full example of downloading files from GitHub Packages, click here.

Why?

I wrote this article because many, many folks have inquired about how to manually download files from OCI registries on the Internet, but their simple queries are usually returned with a barrage of useless counter-questions: why the heck would you want to do that!?!

The answer is varied.

Some people need to get files onto a restricted environment. Either their org doesn't grant them permission to install software on the machine, or the system has firewall-restricted internet access -- or doesn't have internet access at all.

3TOFU

Personally, the reason that I wanted to be able to download files from an OCI registry was for 3TOFU.

Verifying Unsigned Releases with 3TOFU

Unfortunaetly, most apps using OCI registries are extremely insecure. Docker, for example, will happily download malicious images. By default, it doesn't do any authenticity verifications on the payloads it downloaded. Even if you manually enable DCT, there's loads of pending issues with it.

Likewise, the macOS package manager brew has this same problem: it will happily download and install malicious code, because it doesn't use cryptography to verify the authenticity of anything that it downloads. This introduces watering hole vulnerabilities when developers use brew to install dependencies in their CI pipelines.

My solution to this? 3TOFU. And that requires me to be able to download the file (for verification) on three distinct linux VMs using curl or wget.

⚠ NOTE: 3TOFU is an approach to harm reduction.

It is not wise to download and run binaries or code whose authenticity you cannot verify using a cryptographic signature from a key stored offline. However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following a 3TOFU procedure may reduce your risk, but it's better to avoid running unauthenticated code if at all possible.

Registry (ab)use

Container registries were created in 2013 to provide a clever & complex solution to a problem: how to package and serve multiple versions of simplified sources to various consumers spanning multiple operating systems and architectures -- while also packaging them into small, discrete "layers".

However, if your project is just serving simple files, then the only thing gained by uploading them to a complex system like a container registry is headaches. Why do developers do this?

In the case of brew, their free hosing provider (JFrog's Bintray) shutdown in 2021. Brew was already hosting their code on GitHub, so I guess someone looked at "GitHub Packages" and figured it was a good (read: free) replacement.

Many developers using Container Registries don't need the complexity, but -- well -- they're just using it as a free place for their FOSS project to store some files, man.

3TOFU: Verifying Unsigned Releases

By Michael Altfield
License: CC BY-SA 4.0
https://tech.michaelaltfield.net

This article introduces the concept of "3TOFU" -- a harm-reduction process when downloading software that cannot be verified cryptographically.

Verifying Unsigned Releases with 3TOFU

⚠ NOTE: This article is about harm reduction.

It is dangerous to download and run binaries (or code) whose authenticity you cannot verify (using a cryptographic signature from a key stored offline). However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following the steps outlined in this guide may reduce your risk.

TOFU

TOFU stands for Trust On First Use. It's a (often abused) concept of downloading a person or org's signing key and just blindly trusting it (instead of verifying it).

3TOFU

3TOFU is a process where a user downloads something three times at three different locations. If-and-only-if all three downloads are identical, then you trust it.

Why 3TOFU?

During the Crypto Wars of the 1990s, it was illegal to export cryptography from the United States. In 1996, after intense public pressure and legal challenges, the government officially permitted export with the 56-bit DES cipher -- which was a known-vulnerable cipher.

The EFF's Deep Crack proved DES to be insecure and pushed a switch to 3DES.

But there was a simple way to use insecure DES to make secure messages: just use it three times.

3DES (aka "Triple DES") is the process encrypting a message using the insecure symmetric block cipher (DES) three times on each block, to produce an actually secure message (from known attacks at the time).

3TOFU (aka "Triple TOFU") is the process of downloading a payload using the insecure method (TOFU) three times, to obtain the payload that's magnitudes less likely to be maliciously altered.

3TOFU Process

To best mitigate targeted attacks, 3TOFU should be done:

1. On three distinct days
2. On three distinct machines (or VMs)
3. Exiting from three distinct countries
4. Exiting using three distinct networks

For example, I'll usually execute

• TOFU #1/3 in TAILS (via Tor)
• TOFU #2/3 in a Debian VM (via VPN)
• TOFU #3/3 on my daily laptop (via ISP)

The possibility of an attacker maliciously modifying something you download over your ISP's network are quite high, depending on which country you live-in.

The possibility of an attacker maliciously modifying something you download onto a VM with a freshly installed OS over an encrypted VPN connection (routed internationally and exiting from another country) is much less likely, but still possible -- especially for a well-funded adversary.

The possibility of an attacker maliciously modifying something you download onto a VM running a hardened OS (like Whonix or TAILS) using a hardened browser (like Tor Browser) over an anonymizing network (like Tor) is quite unlikely.

The possibility for someone to execute a network attack on all three downloads is very near-zero -- especially if the downloads were spread-out over days or weeks.

3TOFU bash Script

I provide the following bash script as an example snippet that I run for each of the 3TOFUs.

REMOTE_FILES="https://tails.net/tails-signing.key"

CURL="/usr/bin/curl"
WGET="/usr/bin/wget --retry-on-host-error --retry-connrefused"
PYTHON="/usr/bin/python3"

# in tails, we must torify
if [[ "`whoami`" == "amnesia" ]] ; then
	CURL="/usr/bin/torify ${CURL}"
	WGET="/usr/bin/torify ${WGET}"
	PYTHON="/usr/bin/torify ${PYTHON}"
fi

tmpDir=`mktemp -d`
pushd "${tmpDir}"

# first get some info about our internet connection
${CURL} -s https://ifconfig.co/country | head -n1
${CURL} -s https://check.torproject.org | grep Congratulations | head -n1

# and today's date
date -u +"%Y-%m-%d"

# get the file
for file in ${REMOTE_FILES}; do
	wget ${file}
done

# checksum
date -u +"%Y-%m-%d"
sha256sum *

# gpg fingerprint
gpg --with-fingerprint  --with-subkey-fingerprint --keyid-format 0xlong *

Here's one example execution of the above script (on a debian DispVM, executed with a VPN).

/tmp/tmp.xT9HCeTY0y ~
Canada
2024-05-04
--2024-05-04 14:58:54--  https://tails.net/tails-signing.key
Resolving tails.net (tails.net)... 204.13.164.63
Connecting to tails.net (tails.net)|204.13.164.63|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1387192 (1.3M) [application/octet-stream]
Saving to: ‘tails-signing.key’

tails-signing.key   100%[===================>]   1.32M  1.26MB/s    in 1.1s    

2024-05-04 14:58:56 (1.26 MB/s) - ‘tails-signing.key’ saved [1387192/1387192]

2024-05-04
8c641252767dc8815d3453e540142ea143498f8fbd76850066dc134445b3e532  tails-signing.key
gpg: WARNING: no command supplied.  Trying to guess what you mean ...
pub   rsa4096/0xDBB802B258ACD84F 2015-01-18 [C] [expires: 2025-01-25]
      Key fingerprint = A490 D0F4 D311 A415 3E2B  B7CA DBB8 02B2 58AC D84F
uid                             Tails developers (offline long-term identity key) <tails@boum.org>
uid                             Tails developers <tails@boum.org>
sub   rsa4096/0x3C83DCB52F699C56 2015-01-18 [S] [expired: 2018-01-11]
sub   rsa4096/0x98FEC6BC752A3DB6 2015-01-18 [S] [expired: 2018-01-11]
sub   rsa4096/0xAA9E014656987A65 2015-01-18 [S] [revoked: 2015-10-29]
sub   rsa4096/0xAF292B44A0EDAA41 2016-08-30 [S] [expired: 2018-01-11]
sub   rsa4096/0xD21DAD38AF281C0B 2017-08-28 [S] [expires: 2025-01-25]
sub   rsa4096/0x3020A7A9C2B72733 2017-08-28 [S] [revoked: 2020-05-29]
sub   ed25519/0x90B2B4BD7AED235F 2017-08-28 [S] [expires: 2025-01-25]
sub   rsa4096/0xA8B0F4E45B1B50E2 2018-08-30 [S] [revoked: 2021-10-14]
sub   rsa4096/0x7BFBD2B902EE13D0 2021-10-14 [S] [expires: 2025-01-25]
sub   rsa4096/0xE5DBA2E186D5BAFC 2023-10-03 [S] [expires: 2025-01-25]

The TOFU output above shows that the release signing key from the TAILS project is a 4096-bit RSA key with a full fingerprint of "A490 D0F4 D311 A415 3E2B B7CA DBB8 02B2 58AC D84F". The key file itself has a sha256 hash of "8c641252767dc8815d3453e540142ea143498f8fbd76850066dc134445b3e532".

When doing a 3TOFU, save the output of each execution. After collecting output from all 3 executions (intentionally spread-out over 3 days or more), diff the output.

If the output of all three TOFUs match, then the confidence of the file's authenticity is very high.

Why do 3TOFU?

Unfortunately, many developers think that hosting their releases on a server with https is sufficient to protect their users from obtaining a maliciously-modified release. But https won't protect you if:

1. Your DNS or publishing infrastructure is compromised (it happens), or
2. An attacker has just one (subordinate) CA in the user's PKI root store (it happens)

Generally speaking, publishing infrastructure compromises are detected and resolved within days and MITM attacks using compromised CAs are targeted attacks (to avoid detection). Therefore, a 3TOFU verification should thwart these types of attacks.

⚠ Note on hashes: Unfortunately, many well-meaning developers erroneously think that cryptographic hashes provide authenticity, but cryptographic hashes do not provide authenticity -- they provide integrity.

Integrity checks are useful to detect corrupted data on-download; it does not protect you from maliciously altered data unless those hashes are cryptographically signed with a key whose private key isn't stored on the publishing infrastructure.

Improvements

There are some things you can do to further improve the confidence of the authenticity of a file you download from the internet.

Distinct Domains

If possible, download your payload from as many distinct domains as possible.

An adversary may successfully compromise the publishing infrastructure of a software project, but it's far less likely for them to compromise the project website (eg 'tails.net') and their forge (eg 'github.com') and their mastodon instance (eg 'mastodon.social').

Use TAILS

TAILS is by far the best OS to use for security-critical situations.

If you are a high-risk target (investigative journalist, activist, or political dissident) then you should definitely use TAILS for one of your TOFUs.

Signature Verification

It's always better to verify the authenticity of a file using cryptographic signatures than with 3TOFU.

Unfortunately, some companies like Microsoft don't sign their releases, so the only option to verify the authenticity of something like a Windows .iso is with 3TOFU.

Still, whenever you encounter some software that is not signed using an offline key, please do us all a favor and create a bug report asking the developer to sign their releases with PGP (or minisign or signify or something).

4TOFU

3TOFU is easy because Tor is free and most people have access to a VPN (corporate or commercial or an ssh socks proxy).

But, if you'd like, you could also add i2p or some other proxy network into the mix (and do 4TOFU).

3TOFU: Verifying Unsigned Releases

By Michael Altfield
License: CC BY-SA 4.0
https://tech.michaelaltfield.net

This article introduces the concept of "3TOFU" -- a harm-reduction process when downloading software that cannot be verified cryptographically.

Verifying Unsigned Releases with 3TOFU

⚠ NOTE: This article is about harm reduction.

It is dangerous to download and run binaries (or code) whose authenticity you cannot verify (using a cryptographic signature from a key stored offline). However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following the steps outlined in this guide may reduce your risk.

TOFU

TOFU stands for Trust On First Use. It's a (often abused) concept of downloading a person or org's signing key and just blindly trusting it (instead of verifying it).

3TOFU

3TOFU is a process where a user downloads something three times at three different locations. If-and-only-if all three downloads are identical, then you trust it.

Why 3TOFU?

During the Crypto Wars of the 1990s, it was illegal to export cryptography from the United States. In 1996, after intense public pressure and legal challenges, the government officially permitted export with the 56-bit DES cipher -- which was a known-vulnerable cipher.

The EFF's Deep Crack proved DES to be insecure and pushed a switch to 3DES.

But there was a simple way to use insecure DES to make secure messages: just use it three times.

3DES (aka "Triple DES") is the process encrypting a message using the insecure symmetric block cipher (DES) three times on each block, to produce an actually secure message (from known attacks at the time).

3TOFU (aka "Triple TOFU") is the process of downloading a payload using the insecure method (TOFU) three times, to obtain the payload that's magnitudes less likely to be maliciously altered.

3TOFU Process

To best mitigate targeted attacks, 3TOFU should be done:

1. On three distinct days
2. On three distinct machines (or VMs)
3. Exiting from three distinct countries
4. Exiting using three distinct networks

For example, I'll usually execute

• TOFU #1/3 in TAILS (via Tor)
• TOFU #2/3 in a Debian VM (via VPN)
• TOFU #3/3 on my daily laptop (via ISP)

The possibility of an attacker maliciously modifying something you download over your ISP's network are quite high, depending on which country you live-in.

The possibility of an attacker maliciously modifying something you download onto a VM with a freshly installed OS over an encrypted VPN connection (routed internationally and exiting from another country) is much less likely, but still possible -- especially for a well-funded adversary.

The possibility of an attacker maliciously modifying something you download onto a VM running a hardened OS (like Whonix or TAILS) using a hardened browser (like Tor Browser) over an anonymizing network (like Tor) is quite unlikely.

The possibility for someone to execute a network attack on all three downloads is very near-zero -- especially if the downloads were spread-out over days or weeks.

3TOFU bash Script

I provide the following bash script as an example snippet that I run for each of the 3TOFUs.

REMOTE_FILES="https://tails.net/tails-signing.key"

CURL="/usr/bin/curl"
WGET="/usr/bin/wget --retry-on-host-error --retry-connrefused"
PYTHON="/usr/bin/python3"

# in tails, we must torify
if [[ "`whoami`" == "amnesia" ]] ; then
	CURL="/usr/bin/torify ${CURL}"
	WGET="/usr/bin/torify ${WGET}"
	PYTHON="/usr/bin/torify ${PYTHON}"
fi

tmpDir=`mktemp -d`
pushd "${tmpDir}"

# first get some info about our internet connection
${CURL} -s https://ifconfig.co/country | head -n1
${CURL} -s https://check.torproject.org | grep Congratulations | head -n1

# and today's date
date -u +"%Y-%m-%d"

# get the file
for file in ${REMOTE_FILES}; do
	wget ${file}
done

# checksum
date -u +"%Y-%m-%d"
sha256sum *

# gpg fingerprint
gpg --with-fingerprint  --with-subkey-fingerprint --keyid-format 0xlong *

Here's one example execution of the above script (on a debian DispVM, executed with a VPN).

/tmp/tmp.xT9HCeTY0y ~
Canada
2024-05-04
--2024-05-04 14:58:54--  https://tails.net/tails-signing.key
Resolving tails.net (tails.net)... 204.13.164.63
Connecting to tails.net (tails.net)|204.13.164.63|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1387192 (1.3M) [application/octet-stream]
Saving to: ‘tails-signing.key’

tails-signing.key   100%[===================>]   1.32M  1.26MB/s    in 1.1s    

2024-05-04 14:58:56 (1.26 MB/s) - ‘tails-signing.key’ saved [1387192/1387192]

2024-05-04
8c641252767dc8815d3453e540142ea143498f8fbd76850066dc134445b3e532  tails-signing.key
gpg: WARNING: no command supplied.  Trying to guess what you mean ...
pub   rsa4096/0xDBB802B258ACD84F 2015-01-18 [C] [expires: 2025-01-25]
      Key fingerprint = A490 D0F4 D311 A415 3E2B  B7CA DBB8 02B2 58AC D84F
uid                             Tails developers (offline long-term identity key) <tails@boum.org>
uid                             Tails developers <tails@boum.org>
sub   rsa4096/0x3C83DCB52F699C56 2015-01-18 [S] [expired: 2018-01-11]
sub   rsa4096/0x98FEC6BC752A3DB6 2015-01-18 [S] [expired: 2018-01-11]
sub   rsa4096/0xAA9E014656987A65 2015-01-18 [S] [revoked: 2015-10-29]
sub   rsa4096/0xAF292B44A0EDAA41 2016-08-30 [S] [expired: 2018-01-11]
sub   rsa4096/0xD21DAD38AF281C0B 2017-08-28 [S] [expires: 2025-01-25]
sub   rsa4096/0x3020A7A9C2B72733 2017-08-28 [S] [revoked: 2020-05-29]
sub   ed25519/0x90B2B4BD7AED235F 2017-08-28 [S] [expires: 2025-01-25]
sub   rsa4096/0xA8B0F4E45B1B50E2 2018-08-30 [S] [revoked: 2021-10-14]
sub   rsa4096/0x7BFBD2B902EE13D0 2021-10-14 [S] [expires: 2025-01-25]
sub   rsa4096/0xE5DBA2E186D5BAFC 2023-10-03 [S] [expires: 2025-01-25]

The TOFU output above shows that the release signing key from the TAILS project is a 4096-bit RSA key with a full fingerprint of "A490 D0F4 D311 A415 3E2B B7CA DBB8 02B2 58AC D84F". The key file itself has a sha256 hash of "8c641252767dc8815d3453e540142ea143498f8fbd76850066dc134445b3e532".

When doing a 3TOFU, save the output of each execution. After collecting output from all 3 executions (intentionally spread-out over 3 days or more), diff the output.

If the output of all three TOFUs match, then the confidence of the file's authenticity is very high.

Why do 3TOFU?

Unfortunately, many developers think that hosting their releases on a server with https is sufficient to protect their users from obtaining a maliciously-modified release. But https won't protect you if:

1. Your DNS or publishing infrastructure is compromised (it happens), or
2. An attacker has just one (subordinate) CA in the user's PKI root store (it happens)

Generally speaking, publishing infrastructure compromises are detected and resolved within days and MITM attacks using compromised CAs are targeted attacks (to avoid detection). Therefore, a 3TOFU verification should thwart these types of attacks.

⚠ Note on hashes: Unfortunately, many well-meaning developers erroneously think that cryptographic hashes provide authenticity, but cryptographic hashes do not provide authenticity -- they provide integrity.

Integrity checks are useful to detect corrupted data on-download; it does not protect you from maliciously altered data unless those hashes are cryptographically signed with a key whose private key isn't stored on the publishing infrastructure.

Improvements

There are some things you can do to further improve the confidence of the authenticity of a file you download from the internet.

Distinct Domains

If possible, download your payload from as many distinct domains as possible.

An adversary may successfully compromise the publishing infrastructure of a software project, but it's far less likely for them to compromise the project website (eg 'tails.net') and their forge (eg 'github.com') and their mastodon instance (eg 'mastodon.social').

Use TAILS

TAILS is by far the best OS to use for security-critical situations.

If you are a high-risk target (investigative journalist, activist, or political dissident) then you should definitely use TAILS for one of your TOFUs.

Signature Verification

It's always better to verify the authenticity of a file using cryptographic signatures than with 3TOFU.

Unfortunately, some companies like Microsoft don't sign their releases, so the only option to verify the authenticity of something like a Windows .iso is with 3TOFU.

Still, whenever you encounter some software that is not signed using an offline key, please do us all a favor and create a bug report asking the developer to sign their releases with PGP (or minisign or signify or something).

4TOFU

3TOFU is easy because Tor is free and most people have access to a VPN (corporate or commercial or an ssh socks proxy).

But, if you'd like, you could also add i2p or some other proxy network into the mix (and do 4TOFU).

3TOFU: Verifying Unsigned Releases

By Michael Altfield
License: CC BY-SA 4.0
https://tech.michaelaltfield.net

This article introduces the concept of "3TOFU" -- a harm-reduction process when downloading software that cannot be verified cryptographically.

Verifying Unsigned Releases with 3TOFU

⚠ NOTE: This article is about harm reduction.

It is dangerous to download and run binaries (or code) whose authenticity you cannot verify (using a cryptographic signature from a key stored offline). However, sometimes we cannot avoid it. If you're going to proceed with running untrusted code, then following the steps outlined in this guide may reduce your risk.

TOFU

TOFU stands for Trust On First Use. It's a (often abused) concept of downloading a person or org's signing key and just blindly trusting it (instead of verifying it).

3TOFU

3TOFU is a process where a user downloads something three times at three different locations. If-and-only-if all three downloads are identical, then you trust it.

Why 3TOFU?

During the Crypto Wars of the 1990s, it was illegal to export cryptography from the United States. In 1996, after intense public pressure and legal challenges, the government officially permitted export with the 56-bit DES cipher -- which was a known-vulnerable cipher.

The EFF's Deep Crack proved DES to be insecure and pushed a switch to 3DES.

But there was a simple way to use insecure DES to make secure messages: just use it three times.

3DES (aka "Triple DES") is the process encrypting a message using the insecure symmetric block cipher (DES) three times on each block, to produce an actually secure message (from known attacks at the time).

3TOFU (aka "Triple TOFU") is the process of downloading a payload using the insecure method (TOFU) three times, to obtain the payload that's magnitudes less likely to be maliciously altered.

3TOFU Process

To best mitigate targeted attacks, 3TOFU should be done:

1. On three distinct days
2. On three distinct machines (or VMs)
3. Exiting from three distinct countries
4. Exiting using three distinct networks

For example, I'll usually execute

• TOFU #1/3 in TAILS (via Tor)
• TOFU #2/3 in a Debian VM (via VPN)
• TOFU #3/3 on my daily laptop (via ISP)

The possibility of an attacker maliciously modifying something you download over your ISP's network are quite high, depending on which country you live-in.

The possibility of an attacker maliciously modifying something you download onto a VM with a freshly installed OS over an encrypted VPN connection (routed internationally and exiting from another country) is much less likely, but still possible -- especially for a well-funded adversary.

The possibility of an attacker maliciously modifying something you download onto a VM running a hardened OS (like Whonix or TAILS) using a hardened browser (like Tor Browser) over an anonymizing network (like Tor) is quite unlikely.

The possibility for someone to execute a network attack on all three downloads is very near-zero -- especially if the downloads were spread-out over days or weeks.

3TOFU bash Script

I provide the following bash script as an example snippet that I run for each of the 3TOFUs.

REMOTE_FILES="https://tails.net/tails-signing.key"

CURL="/usr/bin/curl"
WGET="/usr/bin/wget --retry-on-host-error --retry-connrefused"
PYTHON="/usr/bin/python3"

# in tails, we must torify
if [[ "`whoami`" == "amnesia" ]] ; then
	CURL="/usr/bin/torify ${CURL}"
	WGET="/usr/bin/torify ${WGET}"
	PYTHON="/usr/bin/torify ${PYTHON}"
fi

tmpDir=`mktemp -d`
pushd "${tmpDir}"

# first get some info about our internet connection
${CURL} -s https://ifconfig.co/country | head -n1
${CURL} -s https://check.torproject.org | grep Congratulations | head -n1

# and today's date
date -u +"%Y-%m-%d"

# get the file
for file in ${REMOTE_FILES}; do
	wget ${file}
done

# checksum
date -u +"%Y-%m-%d"
sha256sum *

# gpg fingerprint
gpg --with-fingerprint  --with-subkey-fingerprint --keyid-format 0xlong *

Here's one example execution of the above script (on a debian DispVM, executed with a VPN).

/tmp/tmp.xT9HCeTY0y ~
Canada
2024-05-04
--2024-05-04 14:58:54--  https://tails.net/tails-signing.key
Resolving tails.net (tails.net)... 204.13.164.63
Connecting to tails.net (tails.net)|204.13.164.63|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1387192 (1.3M) [application/octet-stream]
Saving to: ‘tails-signing.key’

tails-signing.key   100%[===================>]   1.32M  1.26MB/s    in 1.1s    

2024-05-04 14:58:56 (1.26 MB/s) - ‘tails-signing.key’ saved [1387192/1387192]

2024-05-04
8c641252767dc8815d3453e540142ea143498f8fbd76850066dc134445b3e532  tails-signing.key
gpg: WARNING: no command supplied.  Trying to guess what you mean ...
pub   rsa4096/0xDBB802B258ACD84F 2015-01-18 [C] [expires: 2025-01-25]
      Key fingerprint = A490 D0F4 D311 A415 3E2B  B7CA DBB8 02B2 58AC D84F
uid                             Tails developers (offline long-term identity key) <tails@boum.org>
uid                             Tails developers <tails@boum.org>
sub   rsa4096/0x3C83DCB52F699C56 2015-01-18 [S] [expired: 2018-01-11]
sub   rsa4096/0x98FEC6BC752A3DB6 2015-01-18 [S] [expired: 2018-01-11]
sub   rsa4096/0xAA9E014656987A65 2015-01-18 [S] [revoked: 2015-10-29]
sub   rsa4096/0xAF292B44A0EDAA41 2016-08-30 [S] [expired: 2018-01-11]
sub   rsa4096/0xD21DAD38AF281C0B 2017-08-28 [S] [expires: 2025-01-25]
sub   rsa4096/0x3020A7A9C2B72733 2017-08-28 [S] [revoked: 2020-05-29]
sub   ed25519/0x90B2B4BD7AED235F 2017-08-28 [S] [expires: 2025-01-25]
sub   rsa4096/0xA8B0F4E45B1B50E2 2018-08-30 [S] [revoked: 2021-10-14]
sub   rsa4096/0x7BFBD2B902EE13D0 2021-10-14 [S] [expires: 2025-01-25]
sub   rsa4096/0xE5DBA2E186D5BAFC 2023-10-03 [S] [expires: 2025-01-25]

The TOFU output above shows that the release signing key from the TAILS project is a 4096-bit RSA key with a full fingerprint of "A490 D0F4 D311 A415 3E2B B7CA DBB8 02B2 58AC D84F". The key file itself has a sha256 hash of "8c641252767dc8815d3453e540142ea143498f8fbd76850066dc134445b3e532".

When doing a 3TOFU, save the output of each execution. After collecting output from all 3 executions (intentionally spread-out over 3 days or more), diff the output.

If the output of all three TOFUs match, then the confidence of the file's authenticity is very high.

Why do 3TOFU?

Unfortunately, many developers think that hosting their releases on a server with https is sufficient to protect their users from obtaining a maliciously-modified release. But https won't protect you if:

1. Your DNS or publishing infrastructure is compromised (it happens), or
2. An attacker has just one (subordinate) CA in the user's PKI root store (it happens)

Generally speaking, publishing infrastructure compromises are detected and resolved within days and MITM attacks using compromised CAs are targeted attacks (to avoid detection). Therefore, a 3TOFU verification should thwart these types of attacks.

⚠ Note on hashes: Unfortunately, many well-meaning developers erroneously think that cryptographic hashes provide authenticity, but cryptographic hashes do not provide authenticity -- they provide integrity.

Integrity checks are useful to detect corrupted data on-download; it does not protect you from maliciously altered data unless those hashes are cryptographically signed with a key whose private key isn't stored on the publishing infrastructure.

Improvements

There are some things you can do to further improve the confidence of the authenticity of a file you download from the internet.

Distinct Domains

If possible, download your payload from as many distinct domains as possible.

An adversary may successfully compromise the publishing infrastructure of a software project, but it's far less likely for them to compromise the project website (eg 'tails.net') and their forge (eg 'github.com') and their mastodon instance (eg 'mastodon.social').

Use TAILS

TAILS is by far the best OS to use for security-critical situations.

If you are a high-risk target (investigative journalist, activist, or political dissident) then you should definitely use TAILS for one of your TOFUs.

Signature Verification

It's always better to verify the authenticity of a file using cryptographic signatures than with 3TOFU.

Unfortunately, some companies like Microsoft don't sign their releases, so the only option to verify the authenticity of something like a Windows .iso is with 3TOFU.

Still, whenever you encounter some software that is not signed using an offline key, please do us all a favor and create a bug report asking the developer to sign their releases with PGP (or minisign or signify or something).

4TOFU

3TOFU is easy because Tor is free and most people have access to a VPN (corporate or commercial or an ssh socks proxy).

But, if you'd like, you could also add i2p or some other proxy network into the mix (and do 4TOFU).