GHSA-5474-4w2j-mq4c is a medium-severity (CVSS 5.4) Path Traversal vulnerability in @anthropic-ai/sdk. A fix is available for @anthropic-ai/sdk — see the affected versions and patch details below.
Claude SDK for TypeScript: Memory Tool Path Validation Allows Sandbox Escape to Sibling Directories
Exploitation Status
No confirmed exploitation observed yet
- CISA’s own triage has not observed active exploitation or public proof-of-concept code for this CVE as of its last assessment.
Exploitation and automatability from CISA’s SSVC triage for GHSA-5474-4w2j-mq4c.
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
How urgent is this, really
GHSA-5474-4w2j-mq4c plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.
Where this sits among everything scored
Of 378,156 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.
Real-World Exposure
How broadly this vulnerability is actually deployed: weekly install volume shows current usage, and reverse-dependency count shows how many other packages break if it stays unpatched.
@anthropic-ai/sdknpmDescription
Claude SDK for TypeScript provides access to the Claude API from server-side TypeScript or JavaScript applications. From version 0.79.0 to before version 0.81.0, the local filesystem memory tool in the Anthropic TypeScript SDK validated model-supplied paths using a string prefix check that did not append a trailing path separator. A model steered by prompt injection could supply a crafted path that resolved to a sibling directory sharing the memory root's name as a prefix, allowing reads and writes outside the sandboxed memory directory. This issue has been patched in version 0.81.0.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 📦npm | @anthropic-ai/sdk | ≥ 0.79.0&&< 0.81.0 | 0.81.0npm install @anthropic-ai/sdk@0.81.0 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for @anthropic-ai/sdk, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update @anthropic-ai/sdk to 0.81.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-5474-4w2j-mq4c is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like GHSA-5474-4w2j-mq4c can be triaged on real exposure rather than presence alone.
Tailored to GHSA-5474-4w2j-mq4c. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-5474-4w2j-mq4c in your dependencies?
O3 Security finds GHSA-5474-4w2j-mq4c across npm dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.