UTC

onnx

17 published advisories; version 1.22.0 or later fixes all of them.

Registry
PyPI
Advisories
17
Latest fix
1.22.0
Exploited in the wild
None listed

Check your projects

shieldpi agent-config checks agent and MCP packages from npm and PyPI, not onnx. 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/onnx/onnx, named in the package metadata (not verified)

9 of the last 10 releases of onnx 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.

  • 1.23.2
  • 1.23.1
  • 1.23.0
  • 1.22.0
  • 1.21.0
  • 1.20.1
  • 1.20.0
  • 1.19.1
  • 1.19.0
  • 1.18.0

Green: published with provenance. Grey: without.

OpenSSF Scorecard: 8.4 of 10 on 2026-11-17.

No dangerous CI workflows
10/10
CI tokens read-only by default
10/10
Build dependencies pinned
5/10
Branch protection
4/10
Changes reviewed before merge
10/10
Signed releases
Not checked
Actively maintained
10/10
Security policy
10/10
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
ONNX: Out-of-Bounds Read in RNN/GRU/LSTM Shape Inference (Old Opsets, Missing Rank Check)CVE-2026-115919Medium<= 1.21.01.22.02026-10-28
ONNX: Out-of-bounds read in ConvTranspose shape inference via undersized weight rankCVE-2026-115920Medium>= 1.2.1, <= 1.21.01.22.02026-10-28
ONNX: Null Pointer Dereference in ScanInferenceFunction Due to Missing Required AttributeCVE-2026-115921Medium>= 1.4.0, <= 1.21.01.22.02026-10-28
ONNX: Denial of service via unchecked expansion of packed sub-byte tensors in ONNX Python APIsCVE-2026-115922Medium<= 1.21.01.22.02026-10-28
ONNX: Heap-Buffer-Overflow READ in Gemm Version Converter Adapter via Undersized Input ShapeCVE-2026-63632Low>= 1.3.0, <= 1.21.01.22.02026-07-24
ONNX has Null Pointer Dereference in Upsample Version Converter Adapter (Zero Inputs)CVE-2026-44512Medium>= 1.9.0, < 1.22.01.22.02026-07-07
ONNX: Malicious ONNX models can crash servers by exploiting unprotected object settings.CVE-2026-34445High<= 1.20.11.21.02026-04-01
ONNX: Arbitrary File Read via ExternalData Hardlink Bypass in ONNX loadCVE-2026-34446Medium<= 1.20.11.21.02026-04-01
ONNX: External Data Symlink TraversalCVE-2026-34447Medium< 1.21.01.21.02026-04-01
ONNX: TOCTOU arbitrary file read/write in save_external_datCVE-2026-49114High<= 1.20.11.21.02026-04-01
onnx Vulnerable to Path Traversal via SymlinkCVE-2026-27489High<= 1.20.01.21.02026-03-31
ONNX Untrusted Model Repository Warnings Suppressed by silent=True in onnx.hub.load() — Silent Supply-Chain AttackCVE-2026-28500High<= 1.20.11.21.0rc12026-03-16
Open Neural Network Exchange (ONNX) Path Traversal VulnerabilityCVE-2024-7776High< 1.17.01.17.02025-03-20
onnx allows Arbitrary File Overwrite in download_model_with_test_dataCVE-2024-5187High< 1.16.21.16.22024-06-06
Onnx Out-of-bounds Read vulnerabilityCVE-2024-27319Medium<= 1.15.01.16.02024-02-23
Onnx Directory Traversal vulnerabilityCVE-2024-27318High<= 1.15.01.16.02024-02-23
Directory Traversal in onnxCVE-2022-25882High< 1.13.01.13.02023-01-26

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].