Quick answers
- Is CVE-2026-25723 exploited in the wild?
- No. Junglewise Threat Intelligence has no report of CVE-2026-25723 being exploited in the wild.
- How severe is CVE-2026-25723?
- Medium severity, with a CVSS score of 4, published 6 February 2026.
- Which products and versions are affected?
- Anthropic Claude Code < v2.0.55; @anthropic-ai/claude-code (npm) SEMVER:introduced=0,fixed=2.0.55.
- Is there a fix for CVE-2026-25723?
- Yes. @anthropic-ai/claude-code (npm): upgrade to 2.0.55 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
- How is CVE-2026-25723 attacked?
- It is attacked over the network, in a software library.
- Am I affected by CVE-2026-25723?
- Check whether you run Anthropic Claude Code or @anthropic-ai/claude-code (npm) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.
How to fix
- @anthropic-ai/claude-code (npm): upgrade to 2.0.55 or later
npm install @anthropic-ai/claude-code@2.0.55
Technical details
Claude Code failed to properly validate and sanitize commands using piped sed operations combined with the echo command, allowing injection of arbitrary shell commands. An attacker must have the ability to execute commands through Claude Code with the "accept edits" feature enabled (requires user interaction to accept the malicious edits). This vulnerability allows writing to directories that should be protected, bypassing intended file write restrictions. The vulnerability affects versions prior to 2.0.55 and was patched in v2.0.55. Attack vector is network-based via the Claude Code extension.
Affected products
- Anthropic Claude Code < v2.0.55
Identifiers
- CVE-2026-25723
- GHSA-mhg7-666j-cqg4
Timeline
- 2026-02-06: disclosed
- 2026-02-06: patched: Fixed in version 2.0.55