N/A

CVSS Score

-

CVSS Score

Basic Information

Concerned about an active attack path?

Talk to our security experts and see Miggo in action.

Miggo Vulnerability Database

→

CVE-2025-66479

CVE-2025-66479: Anthropic Sandbox Runtime Incorrectly Implemented Network Sandboxing

Due to a bug in sandboxing logic, sandbox-runtime did not properly enforce a network sandbox if the sandbox policy did not configure any allowed domains. This could allow sandboxed code to make network requests outside of the sandbox. A patch for this was released in v0.0.16.

Thank you to https://github.com/bendrucker for reporting this issue!

(GitHub Advisory)

Miggo Vulnerability Database

→

CVE-2025-66479

CVE-2025-66479:

N/A

CVSS Score

-

CVSS Score

Basic Information

Is this CVE running in your environment?

Easily map the attack path and prioritize which CVEs are a threat to your organization

Validate Exposure

Technical Details

Package Name	Ecosystem	Vulnerable Versions	First Patched Version
@anthropic-ai/sandbox-runtime	npm	< 0.0.16	0.0.16

Technical Details

Vulnerability Intelligence
Miggo AI

Root Cause Analysis

The vulnerability was caused by a logical error in the sandbox-manager.ts file within the wrapWithSandbox function. The system was designed to restrict network access based on an allowedDomains list. The documented behavior stated that an empty allowedDomains array should block all network access. However, the code incorrectly equated an empty list with a disabled network sandbox.

The key flaw was the line const needsNetworkProxy = allowedDomains.length > 0, the result of which was then used to determine if network restrictions should be applied (needsNetworkRestriction: needsNetworkProxy). When a user provided an empty allowedDomains: [], needsNetworkProxy became false, which in turn meant needsNetworkRestriction was false. This caused the sandbox to be created without any network isolation, allowing sandboxed code to make unrestricted network requests.

The patch addresses this by decoupling the concept of "network restriction" from "network proxying". A new variable, needsNetworkRestriction, is set to true whenever any network configuration is provided, including an empty allowedDomains array. This ensures that the platform-specific sandboxing functions (wrapCommandWithSandboxLinux and wrapCommandWithSandboxMacOS) are always invoked with the instruction to isolate the network. The fix in wrapCommandWithSandboxLinux then uses --unshare-net to create a new, empty network namespace, effectively blocking all network access unless proxy sockets are explicitly provided for filtered access.

Vulnerable functions

Only Mi**o us*rs **n s** t*is s**tion

Vulnerability Intelligence
Miggo AI

Unlock WAF rules for this CVE

Generate vendor-ready rules for the observed attack patterns, plus reasoning and safe deployment guidance

Get WAF rules

WAF Protection Rules

WAF Rule

W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.W** rul*s *v*il**l* *or Mi**o *ustom*rs only.

Reasoning

*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.*v*il**l* *or Mi**o *ustom*rs only.