DEBIAN-CVE-2026-13898

Use after free in Cast Receiver in Google Chrome prior to 150.0.7871.47 allowed a remote attacker to execute arbitrary code inside a sandbox via a crafted HTML page. Chromium security severity: Medium...

CVE-2026-56377

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

UBUNTU-CVE-2026-56377

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

CVE-2026-56377 ImageMagick - Policy Bypass via Incorrect Path Validation

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

CVE-2026-56377

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

CVE-2026-56377

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

CVE-2026-56377 ImageMagick - Policy Bypass via Incorrect Path Validation

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

CVE-2026-56377 ImageMagick - Policy Bypass via Incorrect Path Validation

ImageMagick before 7.1.2-24 contains an incorrect policy check that allows attackers to create or truncate files disallowed by security policies. Remote attackers can bypass path policy restrictions in sandboxed conversion services to write arbitrary files outside intended boundaries...

GHSA-9J7F-3R4P-PWH6 nono-py vulnerable to authorization bypass / policy confusion

The python API made a restrictive-looking configuration unsafe by default. A caller could configure only reverse- proxy credential routes, put the child in CapabilitySet.proxyonly, and reasonably expect network access to be limited to those routes. Instead, because empty allowedhosts meant...

CVE-2026-33235 AutoGPT: Denial of Service (DoS) via Resource Exhaustion in text templating features

AutoGPT is a workflow automation platform for creating, deploying, and managing continuous artificial intelligence agents. In versions prior to 0.6.52, the Fill Text Template block is vulnerable to a Denial of Service DoS attack. While the backend implements a SandboxedEnvironment to prevent...

EUVD-2026-38809

Apple M1 GPUs retain register file data between compute shader dispatches from different processes. A sandboxed Metal attacker app can run a GPU reader shader that reads stale register values left by a separate sandboxed victim app. In the proof of concept, GPUVictim.app generates a fresh random...

CVE-2026-49269

Apple M1 GPUs retain register file data between compute shader dispatches from different processes. A sandboxed Metal attacker app can run a GPU reader shader that reads stale register values left by a separate sandboxed victim app. In the proof of concept, GPUVictim.app generates a fresh random...

CVE-2026-57280

Jenkins Script Security Plugin 1402.v94c9ce464861 and earlier does not intercept the implicit type casts applied to the elements of typed for-each loops in sandboxed Groovy scripts, allowing attackers able to provide such scripts to invoke arbitrary constructors and bypass the sandbox protection...

CVE-2026-57280

The CVE-2026-57280 affects Jenkins Script Security Plugin (versions up to and including 1402.v94c9ce464861). The issue is that sandboxed Groovy scripts do not intercept implicit type casts in elements of typed for-each loops, which can allow a user-supplied script to invoke arbitrary constructors...

CVE-2026-49269

Apple M1 GPUs retain register file data between compute shader dispatches from different processes. A sandboxed Metal attacker app can run a GPU reader shader that reads stale register values left by a separate sandboxed victim app. In the proof of concept, GPUVictim.app generates a fresh random...

CVE-2026-49269

Apple M1 GPUs expose a cross-process register state leakage: a sandboxed Metal attacker can read stale values from another sandboxed process’s compute shader dispatches, potentially recovering a 128-bit secret that was loaded into GPU registers. In proof-of-concept, a victim app writes a fresh se...

CVE-2026-47135

A flaw was found in vm2, an open-source virtual machine VM sandbox for Node.js. An attacker within the sandbox could exploit incomplete symbol interception and missing security checks to gain control over the host system. This could allow the attacker to execute arbitrary code outside the sandbox...

GHSA-H5JC-78HR-3PC9 Sveltia CMS: Stored XSS in Markdown/RichText preview via unsandboxed same-origin iframe

Impact A stored cross-site scripting XSS vulnerability affected the Markdown/RichText field preview renderer in Sveltia CMS. The DOMPurify sanitization configuration used for Markdown previews explicitly permitted iframe elements without enforcing a sandbox attribute or restricting iframe sources...

Arbitrary Argument Injection

Overview bedrock-agentcore is an An SDK for using Bedrock AgentCore Affected versions of this package are vulnerable to Arbitrary Argument Injection via the installpackages function. An attacker can execute arbitrary commands within the sandboxed environment or redirect package resolution to a...

PraisonAI: Compute-bridged file tools allow shell command injection

Compute-bridged file tools allow shell command injection Summary LocalManagedAgent / SandboxedAgent compute bridging wraps readfile, listfiles, and writefile when a compute provider is attached. The bridge converts those file operations into shell command strings using raw path arguments, then...