Sign in Sign up
JA · EN · FR
SecPulse
Categories
CVE-2026-9856 high CVSS 7.1

Path Traversal in transformers (CVE-2026-9856)

Summary

Path traversal in transformers (CVE-2026-9856). Data can be tampered with by attackers. Exploitable via PreTrainedTokenizerBase. Mitigation: upgrade to 5.10.0 or later.

Original description

The description as published by the source.

A vulnerability in huggingface/transformers versions < 5.10.0 allows an attacker to perform arbitrary file writes via path traversal. The issue resides in the save_pretrained() methods of PreTrainedTokenizerBase and ProcessorMixin, where keys from the chat_template dictionary are used directly as filenames without proper validation. An attacker can exploit this by publishing a malicious Hugging Face Hub repository with a crafted tokenizer_config.json file. When a victim downloads and saves the tokenizer or processor, the attacker-controlled keys can escape the intended save directory, enabling arbitrary file writes with attacker-controlled content. This vulnerability affects multiple processors inheriting from ProcessorMixin, including Idefics, Florence, Gemma, Phi, and Qwen-VL.

Source: NIST National Vulnerability Database NVD GitHub Advisory

Auto-generated summary

Assembled by SecPulse from the source data (CVSS, CWE, affected versions) using templates — no AI involved. Check the original description and references below for exact details.

A vulnerability tracked as CVE-2026-9856 has been found in transformers.
Attackers can target a specific entry point like PreTrainedTokenizerBase over the network to misuse the product.
Data can be tampered with by attackers. CVSS score: 7.1/10.
What to do: upgrade transformers to 5.10.0 or later.
If unsure, ask your IT team or search "transformers CVE-2026-9856" on the vendor's site.

CVE-2026-9856 (transformers) — CWE-22 / CVSS v3 7.1
Attack vector: remote (network-reachable) / unauthenticated
Attack surface: PreTrainedTokenizerBase / ProcessorMixin / chat_template
Patched: 5.10.0 — apply immediately
Plan: 1) Audit SBOM/dependencies, 2) Stage→prod upgrade, 3) Add WAF/proxy monitoring on affected endpoints, 4) Hunt IOCs in logs.
Refs: see the GHSA / vendor advisory / patched release linked on this page.

❓ What is the problem

Path traversal (CWE-22) exists in transformers. Attackers reach the vulnerable code path via PreTrainedTokenizerBase without authentication.

📍 Affected scope

Check the vendor advisory for the affected versions of transformers. Attack surface: PreTrainedTokenizerBase / ProcessorMixin / chat_template.

🔥 Severity

Severity: High (CVSS 7.1/10). Data can be tampered with by attackers.

🔧 How to fix

Update to 5.10.0.

🛡️ Workaround

Until the patch is applied: disable the affected feature, apply WAF rules, or restrict access via network ACLs.

🔍 Detection

Search webserver/proxy logs for unusual request patterns matching this CVE's known IOCs. Run grep -r 'transformers' . against your dependency files (package-lock.json, requirements.txt, go.sum) to find affected services.

Response actions

Concrete steps and command examples for SOC/SRE teams to execute in order

  1. 1
    Identify exposure identify 
    grep -r 'transformers' . | grep -v node_modules

    リポジトリと本番環境の依存ファイル (package-lock.json / requirements.txt / go.sum / Gemfile.lock 等) で transformers を grep し、稼働しているサービス・バージョンを把握する。

  2. 2
    Apply patch patch 
    Upgrade transformers to 5.10.0

    ステージング環境で 5.10.0 に上げて回帰テスト → 本番反映。回帰テストはアプリの主要ハッピーパスを含めること。

  3. 3
    Post-deployment verification verify 
    Confirm patched version is live in production

    パッチ適用後、ステージングで PoC または同等の悪用パターンを再現して脆弱性が閉じたことを確認。本番ではアクセスログ・WAF でアラート再発が無いか継続監視。

Affected packages

pip transformers
[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"5.10.0"}]}]

References

We use cookies for analytics and ad measurement. This setting applies to all Game co., Ltd. sites. Details