Packages / @anthropic-ai/sdk / Security
Claude SDK for TypeScript has Insecure Default File Permissions in Local Filesystem Memory Tool
Affected releases · @anthropic-ai/sdk
- 0.91.x0.91.0Fixed in 0.91.1
- 0.90.x0.90.0No fix on this line yet
- 0.89.x0.89.0No fix on this line yet
- 0.88.x0.88.0No fix on this line yet
- 0.87.x0.87.0No fix on this line yet
- 0.86.x0.86.0 – 0.86.1No fix on this line yet
- 0.85.x0.85.0No fix on this line yet
- 0.84.x0.84.0No fix on this line yet
- 0.82.x0.82.0No fix on this line yet
- 0.81.x0.81.0No fix on this line yet
- 0.80.x0.80.0No fix on this line yet
- 0.79.x0.79.0No fix on this line yet
On 0.91.x, the newest line:
npm install @anthropic-ai/sdk@0.91.1Not affected
1.0.0 – 1.1.0
2 releases, latest
Not affected
0.2.0 – 0.78.0
143 releases
Affected
0.79.0 – 0.91.0
13 releases
Fixed
0.91.1 – 0.137.0
61 releases
Is your project affected?
Paste your package.json, or type the range you declare for @anthropic-ai/sdk. It is read in your browser and sent nowhere.
What happens
The BetaLocalFilesystemMemoryTool in the Anthropic TypeScript SDK created memory files and directories using the Node.js default modes (0o666 for files, 0o777 for directories), leaving them world-readable on systems with a standard umask and world-writable in environments with a permissive umask such as many Docker base images. A local attacker on a shared host could read persisted agent state, and in containerized deployments could modify memory files to influence subsequent model behavior.
Users on the affected versions are advised to update to the latest version.
Claude SDK thanks lucasfutures for the report.