This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-45185

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2026-45185

2. CVE-2025-68613

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2025-68613

3. CVE-2021-40539

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2021-40539

4. CVE-2022-21227

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2022-21227

5. CVE-2024-34102

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2024-34102

6. CVE-2024-38063

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2024-38063

7. CVE-2024-49113

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2024-49113

8. CVE-2025-0368

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2025-0368

9. CVE-2025-14847

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2025-14847

10. CVE-2025-21204

Adobe Commerce versions 2.4.7, 2.4.6-p5, 2.4.5-p7, 2.4.4-p8 and earlier are affected by an Improper Restriction of XML External Entity Reference (‘XXE’) vulnerability that could result in arbitrary code execution. An attacker could exploit this vulnerability by sending a crafted XML document that references external entities. Exploitation of this issue does not require user interaction.

https://nvd.nist.gov/vuln/detail/CVE-2025-21204

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-45185

Exim before 4.99.3, in certain GnuTLS configurations, has a remotely reachable use-after-free in the BDAT body parsing path. It is triggered when a client sends a TLS close_notify mid-body during a CHUNKING transfer, followed by a final cleartext byte on the same TCP connection. This can lead to heap corruption. An unauthenticated network attacker exploiting this vulnerability could execute arbitrary code.

https://nvd.nist.gov/vuln/detail/CVE-2026-45185

2. CVE-2025-54100

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2025-54100

3. CVE-2024-34102

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2024-34102

4. CVE-2024-38063

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2024-38063

5. CVE-2024-49113

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2024-49113

6. CVE-2025-0368

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2025-0368

7. CVE-2025-14847

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2025-14847

8. CVE-2025-21204

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2025-21204

9. CVE-2025-24118

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2025-24118

10. CVE-2025-50708

The package sqlite3 before 5.0.3 are vulnerable to Denial of Service (DoS) which will invoke the toString function of the passed parameter. If passed an invalid Function object it will throw and crash the V8 engine.

https://nvd.nist.gov/vuln/detail/CVE-2025-50708

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2024-34102

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2024-34102

2. CVE-2026-31431

In the Linux kernel, the following vulnerability has been resolved:

xfrm: esp: avoid in-place decrypt on shared skb frags

MSG_SPLICE_PAGES can attach pages from a pipe directly to an skb. TCP
marks such skbs with SKBFL_SHARED_FRAG after skb_splice_from_iter(),
so later paths that may modify packet data can first make a private
copy. The IPv4/IPv6 datagram append paths did not set this flag when
splicing pages into UDP skbs.

That leaves an ESP-in-UDP packet made from shared pipe pages looking
like an ordinary uncloned nonlinear skb. ESP input then takes the no-COW
fast path for uncloned skbs without a frag_list and decrypts in place
over data that is not owned privately by the skb.

Mark IPv4/IPv6 datagram splice frags with SKBFL_SHARED_FRAG, matching
TCP. Also make ESP input fall back to skb_cow_data() when the flag is
present, so ESP does not decrypt externally backed frags in place.
Private nonlinear skb frags still use the existing fast path.

This intentionally does not change ESP output. In esp_output_head(),
the path that appends the ESP trailer to existing skb tailroom without
calling skb_cow_data() is not reachable for nonlinear skbs:
skb_tailroom() returns zero when skb->data_len is nonzero, while ESP
tailen is positive. Thus ESP output will either use the separate
destination-frag path or fall back to skb_cow_data().

https://nvd.nist.gov/vuln/detail/CVE-2026-31431

3. CVE-2025-43520

A memory corruption issue was addressed with improved memory handling. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may be able to cause unexpected system termination or write kernel memory.

https://nvd.nist.gov/vuln/detail/CVE-2025-43520

4. CVE-2025-43510

A memory corruption issue was addressed with improved memory handling. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may be able to cause unexpected system termination or write kernel memory.

https://nvd.nist.gov/vuln/detail/CVE-2025-43510

5. CVE-2017-8759

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2017-8759

6. CVE-2025-8088

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2025-8088

7. CVE-2025-68613

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2025-68613

8. CVE-2025-59528

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2025-59528

9. CVE-2025-59287

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2025-59287

10. CVE-2025-58434

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2025-58434

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-31431

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2026-31431

2. CVE-2024-34102

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2024-34102

3. CVE-2016-10033

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2016-10033

4. CVE-2017-8759

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2017-8759

5. CVE-2021-40539

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2021-40539

6. CVE-2022-21227

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2022-21227

7. CVE-2024-38063

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2024-38063

8. CVE-2024-49113

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2024-49113

9. CVE-2025-0368

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2025-0368

10. CVE-2025-14847

Mismatched length fields in Zlib compressed protocol headers may allow a read of uninitialized heap memory by an unauthenticated client. This issue affects all MongoDB Server v7.0 prior to 7.0.28 versions, MongoDB Server v8.0 versions prior to 8.0.17, MongoDB Server v8.2 versions prior to 8.2.3, MongoDB Server v6.0 versions prior to 6.0.27, MongoDB Server v5.0 versions prior to 5.0.32, MongoDB Server v4.4 versions prior to 4.4.30, MongoDB Server v4.2 versions greater than or equal to 4.2.0, MongoDB Server v4.0 versions greater than or equal to 4.0.0, and MongoDB Server v3.6 versions greater than or equal to 3.6.0.

https://nvd.nist.gov/vuln/detail/CVE-2025-14847

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-26322

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-26322

2. CVE-2026-24763

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-24763

3. CVE-2026-25157

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25157

4. CVE-2026-25253

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25253

5. CVE-2026-25475

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25475

6. CVE-2026-26329

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-26329

7. CVE-2026-27001

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-27001

8. CVE-2026-31431

In the Linux kernel, the following vulnerability has been resolved:

crypto: algif_aead – Revert to operating out-of-place

This mostly reverts commit 72548b093ee3 except for the copying of
the associated data.

There is no benefit in operating in-place in algif_aead since the
source and destination come from different mappings. Get rid of
all the complexity added for in-place operation and just copy the
AD directly.

https://nvd.nist.gov/vuln/detail/CVE-2026-31431

9. CVE-2026-27940

llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gguf.cpp is vulnerable to an Integer overflow, leading to an undersized heap allocation. Using the subsequent fread() writes 528+ bytes of attacker-controlled data past the buffer boundary. This is a bypass of a similar bug in the same file – CVE-2025-53630, but the fix overlooked some areas. This vulnerability is fixed in b8146.

https://nvd.nist.gov/vuln/detail/CVE-2026-27940

10. CVE-2025-53630

llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gguf.cpp is vulnerable to an Integer overflow, leading to an undersized heap allocation. Using the subsequent fread() writes 528+ bytes of attacker-controlled data past the buffer boundary. This is a bypass of a similar bug in the same file – CVE-2025-53630, but the fix overlooked some areas. This vulnerability is fixed in b8146.

https://nvd.nist.gov/vuln/detail/CVE-2025-53630

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-25253

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25253

2. CVE-2026-25157

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25157

3. CVE-2026-24763

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-24763

4. CVE-2026-25475

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25475

5. CVE-2026-26322

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-26322

6. CVE-2026-27001

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-27001

7. CVE-2026-26329

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-26329

8. CVE-2025-43510

A memory corruption issue was addressed with improved memory handling. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may be able to cause unexpected system termination or write kernel memory.

https://nvd.nist.gov/vuln/detail/CVE-2025-43510

9. CVE-2026-20700

A memory corruption issue was addressed with improved memory handling. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may be able to cause unexpected system termination or write kernel memory.

https://nvd.nist.gov/vuln/detail/CVE-2026-20700

10. CVE-2025-43529

A memory corruption issue was addressed with improved memory handling. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may be able to cause unexpected system termination or write kernel memory.

https://nvd.nist.gov/vuln/detail/CVE-2025-43529

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-26329

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-26329

2. CVE-2026-26322

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-26322

3. CVE-2026-27001

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-27001

4. CVE-2026-25475

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-25475

5. CVE-2026-24763

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-24763

6. CVE-2026-25253

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-25253

7. CVE-2026-25157

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-25157

8. CVE-2024-38476

Vulnerability in core of Apache HTTP Server 2.4.59 and earlier are vulnerably to information disclosure, SSRF or local script execution via backend applications whose response headers are malicious or exploitable.

Users are recommended to upgrade to version 2.4.60, which fixes this issue.

https://nvd.nist.gov/vuln/detail/CVE-2024-38476

9. CVE-2025-43529

A memory corruption issue was addressed with improved lock state checking. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may cause unexpected changes in memory shared between processes.

https://nvd.nist.gov/vuln/detail/CVE-2025-43529

10. CVE-2025-43520

A memory corruption issue was addressed with improved lock state checking. This issue is fixed in iOS 18.7.2 and iPadOS 18.7.2, iOS 26.1 and iPadOS 26.1, macOS Sequoia 15.7.2, macOS Sonoma 14.8.2, macOS Tahoe 26.1, tvOS 26.1, visionOS 26.1, watchOS 26.1. A malicious application may cause unexpected changes in memory shared between processes.

https://nvd.nist.gov/vuln/detail/CVE-2025-43520

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-24763

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-24763

2. CVE-2026-25157

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-25157

3. CVE-2026-25253

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-25253

4. CVE-2026-27001

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-27001

5. CVE-2026-26329

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-26329

6. CVE-2026-26322

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-26322

7. CVE-2026-25475

OpenClaw is a personal AI assistant. Prior to version 2026.2.14, authenticated attackers can read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool’s `upload` action. The server passed these paths to Playwright’s `setInputFiles()` APIs without restricting them to a safe root. An attacker must reach the Gateway HTTP surface (or otherwise invoke the same browser control hook endpoints); present valid Gateway auth (bearer token / password), as required by the Gateway configuration (In common default setups, the Gateway binds to loopback and the onboarding wizard generates a gateway token even for loopback); and have the `browser` tool permitted by tool policy for the target session/context (and have browser support enabled). If an operator exposes the Gateway beyond loopback (LAN/tailnet/custom bind, reverse proxy, tunnels, etc.), the impact increases accordingly. Starting in version 2026.2.14, the upload paths are now confined to OpenClaw’s temp uploads root (`DEFAULT_UPLOAD_DIR`) and traversal/escape paths are rejected.

https://nvd.nist.gov/vuln/detail/CVE-2026-25475

8. CVE-2025-31277

The issue was addressed with improved memory handling. This issue is fixed in Safari 18.6, watchOS 11.6, visionOS 2.6, iOS 18.6 and iPadOS 18.6, macOS Sequoia 15.6, tvOS 18.6. Processing maliciously crafted web content may lead to memory corruption.

https://nvd.nist.gov/vuln/detail/CVE-2025-31277

9. CVE-2025-43520

The issue was addressed with improved memory handling. This issue is fixed in Safari 18.6, watchOS 11.6, visionOS 2.6, iOS 18.6 and iPadOS 18.6, macOS Sequoia 15.6, tvOS 18.6. Processing maliciously crafted web content may lead to memory corruption.

https://nvd.nist.gov/vuln/detail/CVE-2025-43520

10. CVE-2026-20700

The issue was addressed with improved memory handling. This issue is fixed in Safari 18.6, watchOS 11.6, visionOS 2.6, iOS 18.6 and iPadOS 18.6, macOS Sequoia 15.6, tvOS 18.6. Processing maliciously crafted web content may lead to memory corruption.

https://nvd.nist.gov/vuln/detail/CVE-2026-20700

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-32597

Exiv2 is a C++ library and a command-line utility to read, write, delete and modify Exif, IPTC, XMP and ICC image metadata. Prior to version 0.28.8, an uncaught exception was found in Exiv2. The vulnerability is in the preview component, which is only triggered when running Exiv2 with an extra command line argument, like -pp. Due to an integer overflow, the code attempts to create a huge std::vector, which causes Exiv2 to crash with an uncaught exception. This issue has been patched in version 0.28.8.

https://nvd.nist.gov/vuln/detail/CVE-2026-32597

2. CVE-2026-27631

Exiv2 is a C++ library and a command-line utility to read, write, delete and modify Exif, IPTC, XMP and ICC image metadata. Prior to version 0.28.8, an uncaught exception was found in Exiv2. The vulnerability is in the preview component, which is only triggered when running Exiv2 with an extra command line argument, like -pp. Due to an integer overflow, the code attempts to create a huge std::vector, which causes Exiv2 to crash with an uncaught exception. This issue has been patched in version 0.28.8.

https://nvd.nist.gov/vuln/detail/CVE-2026-27631

3. CVE-2025-61147

Exiv2 is a C++ library and a command-line utility to read, write, delete and modify Exif, IPTC, XMP and ICC image metadata. Prior to version 0.28.8, an uncaught exception was found in Exiv2. The vulnerability is in the preview component, which is only triggered when running Exiv2 with an extra command line argument, like -pp. Due to an integer overflow, the code attempts to create a huge std::vector, which causes Exiv2 to crash with an uncaught exception. This issue has been patched in version 0.28.8.

https://nvd.nist.gov/vuln/detail/CVE-2025-61147

4. CVE-2026-3606

Exiv2 is a C++ library and a command-line utility to read, write, delete and modify Exif, IPTC, XMP and ICC image metadata. Prior to version 0.28.8, an uncaught exception was found in Exiv2. The vulnerability is in the preview component, which is only triggered when running Exiv2 with an extra command line argument, like -pp. Due to an integer overflow, the code attempts to create a huge std::vector, which causes Exiv2 to crash with an uncaught exception. This issue has been patched in version 0.28.8.

https://nvd.nist.gov/vuln/detail/CVE-2026-3606

5. CVE-2026-25884

ImageMagick is free and open-source software used for editing and manipulating digital images. Prior to version 7.1.2-12, Magick fails to check for circular references between two MVGs, leading to a stack overflow. This is a DoS vulnerability, and any situation that allows reading the mvg file will be affected. Version 7.1.2-12 fixes the issue.

https://nvd.nist.gov/vuln/detail/CVE-2026-25884

6. CVE-2026-32239

Cap’n Proto is a data interchange format and capability-based RPC system. Prior to 1.4.0, when using Transfer-Encoding: chunked, if a chunk’s size parsed to a value of 2^64 or larger, it would be truncated to a 64-bit integer. In theory, this bug could enable HTTP request/response smuggling. This vulnerability is fixed in 1.4.0.

https://nvd.nist.gov/vuln/detail/CVE-2026-32239

7. CVE-2026-32240

Cap’n Proto is a data interchange format and capability-based RPC system. Prior to 1.4.0, when using Transfer-Encoding: chunked, if a chunk’s size parsed to a value of 2^64 or larger, it would be truncated to a 64-bit integer. In theory, this bug could enable HTTP request/response smuggling. This vulnerability is fixed in 1.4.0.

https://nvd.nist.gov/vuln/detail/CVE-2026-32240

8. CVE-2026-23925

VideoLAN VLC for Android prior to version 3.7.0 contains a path traversal vulnerability in the Remote Access Server routing for the authenticated endpoint GET /download. The file query parameter is concatenated into a filesystem path under the configured download directory without canonicalization or directory containment checks, allowing an authenticated attacker with network reachability to the Remote Access Server to request files outside the intended directory. The impact is bounded by the Android application sandbox and storage restrictions, typically limiting exposure to app-internal and app-specific external storage.

https://nvd.nist.gov/vuln/detail/CVE-2026-23925

9. CVE-2026-27596

ImageMagick is free and open-source software used for editing and manipulating digital images. Prior to version 7.1.2-12, Magick fails to check for circular references between two MVGs, leading to a stack overflow. This is a DoS vulnerability, and any situation that allows reading the mvg file will be affected. Version 7.1.2-12 fixes the issue.

https://nvd.nist.gov/vuln/detail/CVE-2026-27596

10. CVE-2026-31958

PyJWT is a JSON Web Token implementation in Python. Prior to 2.12.0, PyJWT does not validate the crit (Critical) Header Parameter defined in RFC 7515 §4.1.11. When a JWS token contains a crit array listing extensions that PyJWT does not understand, the library accepts the token instead of rejecting it. This violates the MUST requirement in the RFC. This vulnerability is fixed in 2.12.0.

https://nvd.nist.gov/vuln/detail/CVE-2026-31958

This weekly blog post is from via our unique intelligence collection pipelines. We are your eyes and ears online, including the Dark Web.

There are thousands of vulnerability discussions each week. SOS Intelligence gathers a list of the most discussed Common Vulnerabilities and Exposures (CVE) online for the previous week.

We make every effort to ensure the accuracy of the data presented. As this is an automated process some errors may creep in.

If you are feeling generous please do make us aware of anything you spot, feel free to follow us on Twitter @sosintel and DM us. Thank you!

1.  CVE-2026-24763

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-24763

2. CVE-2026-25157

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25157

3. CVE-2026-25253

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25253

4. CVE-2026-25475

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-25475

5. CVE-2026-26322

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-26322

6. CVE-2026-26329

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-26329

7. CVE-2026-27001

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=… would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

https://nvd.nist.gov/vuln/detail/CVE-2026-27001

8. CVE-2024-38476

Vulnerability in core of Apache HTTP Server 2.4.59 and earlier are vulnerably to information disclosure, SSRF or local script execution via backend applications whose response headers are malicious or exploitable.

Users are recommended to upgrade to version 2.4.60, which fixes this issue.

https://nvd.nist.gov/vuln/detail/CVE-2024-38476

9. CVE-2023-52658

In the Linux kernel, the following vulnerability has been resolved:

bpf: Reject narrower access to pointer ctx fields

The following BPF program, simplified from a syzkaller repro, causes a
kernel warning:

r0 = *(u8 *)(r1 + 169);
exit;

With pointer field sk being at offset 168 in __sk_buff. This access is
detected as a narrower read in bpf_skb_is_valid_access because it
doesn’t match offsetof(struct __sk_buff, sk). It is therefore allowed
and later proceeds to bpf_convert_ctx_access. Note that for the
“is_narrower_load” case in the convert_ctx_accesses(), the insn->off
is aligned, so the cnt may not be 0 because it matches the
offsetof(struct __sk_buff, sk) in the bpf_convert_ctx_access. However,
the target_size stays 0 and the verifier errors with a kernel warning:

verifier bug: error during ctx access conversion(1)

This patch fixes that to return a proper “invalid bpf_context access
off=X size=Y” error on the load instruction.

The same issue affects multiple other fields in context structures that
allow narrow access. Some other non-affected fields (for sk_msg,
sk_lookup, and sockopt) were also changed to use bpf_ctx_range_ptr for
consistency.

Note this syzkaller crash was reported in the “Closes” link below, which
used to be about a different bug, fixed in
commit fce7bd8e385a (“bpf/verifier: Handle BPF_LOAD_ACQ instructions
in insn_def_regno()”). Because syzbot somehow confused the two bugs,
the new crash and repro didn’t get reported to the mailing list.

https://nvd.nist.gov/vuln/detail/CVE-2023-52658

10. CVE-2023-53421

In the Linux kernel, the following vulnerability has been resolved:

bpf: Reject narrower access to pointer ctx fields

The following BPF program, simplified from a syzkaller repro, causes a
kernel warning:

r0 = *(u8 *)(r1 + 169);
exit;

With pointer field sk being at offset 168 in __sk_buff. This access is
detected as a narrower read in bpf_skb_is_valid_access because it
doesn’t match offsetof(struct __sk_buff, sk). It is therefore allowed
and later proceeds to bpf_convert_ctx_access. Note that for the
“is_narrower_load” case in the convert_ctx_accesses(), the insn->off
is aligned, so the cnt may not be 0 because it matches the
offsetof(struct __sk_buff, sk) in the bpf_convert_ctx_access. However,
the target_size stays 0 and the verifier errors with a kernel warning:

verifier bug: error during ctx access conversion(1)

This patch fixes that to return a proper “invalid bpf_context access
off=X size=Y” error on the load instruction.

The same issue affects multiple other fields in context structures that
allow narrow access. Some other non-affected fields (for sk_msg,
sk_lookup, and sockopt) were also changed to use bpf_ctx_range_ptr for
consistency.

Note this syzkaller crash was reported in the “Closes” link below, which
used to be about a different bug, fixed in
commit fce7bd8e385a (“bpf/verifier: Handle BPF_LOAD_ACQ instructions
in insn_def_regno()”). Because syzbot somehow confused the two bugs,
the new crash and repro didn’t get reported to the mailing list.

https://nvd.nist.gov/vuln/detail/CVE-2023-53421