MediumVulnerabilityLLM-specific
GHSA-p7fg-763f-g4gf: Claude SDK for TypeScript has Insecure Default File Permissions in Local Filesystem Memory Tool
- Identifiers
- CVE-2026-41686GHSA-p7fg-763f-g4gf
- Published
- Record updated
- Affected
- @anthropic-ai/sdk >= 0.79.0, < 0.91.1
- Fixed in
- 0.91.1
- released April 24, 2026, 5 days before disclosure
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
- EPSS
- 0.1%
Summary
The Anthropic TypeScript SDK's `BetaLocalFilesystemMemoryTool` created memory files and directories with the Node.js default modes (`0o666` for files, `0o777` for directories). On hosts with a standard umask these are world-readable, and under a permissive umask, such as many Docker base images, they are world-writable. A local attacker on a shared host could read persisted agent state, and in containers could modify memory files to influence later model behavior.
Mitigation
Users on the affected versions are advised to update to the latest version.
Affected packages in the Exposure Registry
Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.
- @anthropic-ai/sdknpmLLM dependency since 2023-01-31 · 3 tracked dependents
Related items
- HighCVE-2026-88968: Claude Code Edit tool symlink race overwrites files outside workspace on macOSSame vendor · NVD/CVE Database
- HighCVE-2026-88066: Claude Desktop on Windows git.exe execution via search pathSame vendor · NVD/CVE Database
- HighCVE-2026-62331: praktor path traversal via Telegram document upload filenamesSame vendor · NVD/CVE Database
- MediumGHSA-w4fh-qvv9-3v23: Model Context Protocol Python SDK: Server bearer-token authentication accepts tokens issued for a different resource serverSame vendor · GitHub Advisory Database
- MediumGHSA-5h93-6whr-6q8j: Model Context Protocol Python SDK: HTTP client transports followed cross-origin redirects carrying custom headers and bodiesSame vendor · GitHub Advisory Database