UTC

ray

11 published advisories; version 2.56.0 or later fixes all but 2 of them.

Registry
PyPI
Advisories
11
Latest fix
2.56.0
Exploited in the wild
1
Advisories with no fix yet

2 advisories have no fixed release in any version yet. What can be done meanwhile:

General guidance from ShieldPi, not a vendor fix

  • Command or code injection. Remove or disable the affected server or tool, or run it in a sandbox with no secrets in its environment.GHSA-6wgj-66m2-xxp2
  • Server-side request forgery. Restrict its outbound network access to the hosts it needs, and keep internal addresses and cloud metadata endpoints out of its reach.GHSA-6wgj-66m2-xxp2
  • Missing or bypassable authentication. Do not expose it on a network others can reach, and put authentication you control in front of it, such as an authenticating proxy or a VPN. Binding it to localhost is not enough on its own: a web page open in your browser can still reach it, so require a token as well.GHSA-gx77-xgc2-4888

Check your projects

shieldpi agent-config checks agent and MCP packages from npm and PyPI, not ray. Any scanner that reads the OSV database finds it in your lockfiles, for example Google's open-source OSV-Scanner.

osv-scanner scan source -r .

Or check a lockfile against ShieldMap in your browser.

Provenance

Source: github.com/ray-project/ray, named in the package metadata (not verified)

0 of the last 10 releases of ray were published with build provenance: a signed record of the repository and commit each was built from, which the registry checks before it accepts the release.

  • 2.59.1
  • 2.60.0
  • 2.58.1
  • 2.57.1
  • 2.59.0
  • 2.58.0
  • 2.57.0
  • 2.56.1
  • 2.56.0
  • 2.55.1

Green: published with provenance. Grey: without.

OpenSSF Scorecard: 5.7 of 10 on 2026-10-19.

No dangerous CI workflows
10/10
CI tokens read-only by default
0/10
Build dependencies pinned
0/10
Branch protection
8/10
Changes reviewed before merge
9/10
Signed releases
Not checked
Actively maintained
10/10
Security policy
3/10
Full Scorecard →

Provenance from deps.dev (CC BY 4.0); Scorecard from OpenSSF (CDLA-Permissive-2.0). As of 2026-11-18.

Advisories
AdvisorySeverityAffectedFixed inPublished
Ray: Arbitrary code execution via ray.data.read_webdataset default decoder: pickle.loads(value) and torch.load(weights_only=False)CVE-2026-57516High< 2.56.02.56.02026-07-24
Ray: Remote Code Execution via Parquet Arrow Extension Type DeserializationCVE-2026-41486High>= 2.49.0, < 2.55.02.55.02026-04-24
Ray Dashboard is vulnerable to path traversal through its static file handling mechanismCVE-2026-32981High< 2.8.12.8.12026-03-17
Ray dashboard DELETE endpoints allow unauthenticated browser-triggered DoS (Serve shutdown / job deletion)CVE-2026-27482Medium< 2.54.02.54.02026-02-20
Ray's New Token Authentication is Disabled By DefaultCVE-2025-34351Critical<= 2.52.0No fix yet2025-11-27
Ray is vulnerable to Critical RCE via Safari & Firefox Browsers through DNS Rebinding AttackCVE-2025-62593Exploited in the wildCritical< 2.52.02.52.02025-11-26
ray vulnerable to Insertion of Sensitive Information into Log FileCVE-2025-1979Medium< 2.43.02.43.02025-03-06
Ray has arbitrary code execution via jobs submission APICVE-2023-48022Critical<= 2.49.2No fix yet2023-11-28
Ray Path Traversal vulnerabilityCVE-2023-6021Critical< 2.8.12.8.12023-11-16
Ray Missing Authorization vulnerabilityCVE-2023-6020Critical< 2.8.12.8.12023-11-16
Ray OS Command Injection vulnerabilityCVE-2023-6019Critical< 2.8.12.8.12023-11-16

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].