Browse / Agent Context and Memory Poisoning / CVE-2026-34450
Claude SDK for Python has Insecure Default File Permissions in Local Filesystem Memory Tool
- Class
- Agent Context and Memory Poisoning (filed by a model that read the advisory)
- Technique
- Information exposure
- Also involves
- Sensitive Data Disclosure, AI Framework and Platform Flaws
- Affected
- anthropic (PyPI)
- CVSS
- CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N
- EPSS
- 0.1% chance of exploitation in 30 days
How it works, in short
The memory tool writes persisted agent memory files with world-readable and world-writable permissions, letting local users read or alter them to steer later model behaviour. (Our summary, written by a model. The source text is below.)
Details
The local filesystem memory tool in the Anthropic Python SDK created memory files with mode 0o666, 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. Both the synchronous and asynchronous memory tool implementations were affected.
Users on the affected versions are advised to update to the latest version.
Claude SDK for Python thanks lucasfutures on HackerOne for the report.
References
- https://github.com/anthropics/anthropic-sdk-python/security/advisories/GHSA-q5f5-3gjm-7mfm
- https://nvd.nist.gov/vuln/detail/CVE-2026-34450
- https://github.com/anthropics/anthropic-sdk-python/commit/715030ceb4d6dd8d3546e999c680e29532bf1255
- https://github.com/anthropics/anthropic-sdk-python
- https://github.com/anthropics/anthropic-sdk-python/releases/tag/v0.87.0
Source: GitHub Advisory Database, used under CC BY 4.0.