Cpython: [CVE-2026-82049] tarfile extraction filters allow file modification and content disclosure via hard link to symlink
Python discloses CVE-2026-82049: tarfile extraction filters can be bypassed via hard links to symlinks, enabling file modification and content disclosure.
Stan Ulbrych disclosed CVE-2026-82049 on the Python security-announce list, forwarded to oss-security by Alan Coopersmith. The flaw affects CPython's tarfile extraction filters, which can allow file modification and content disclosure when archives contain a hard link to a symlink. Extraction-filter misuse can therefore bypass intended protections during archive unpacking.
CVE-2026-86792: Apache Airflow Apache Kafka provider: Connection-editor remote code execution on the Scheduler via Kafka connection callback configuration
Apache Airflow Kafka provider 1.15.0 before 2.0.0 resolves unvalidated dotted-path strings into Python callables, enabling Scheduler RCE.
CVE-2026-86792 (moderate) affects apache-airflow-providers-apache-kafka versions 1.15.0 before 2.0.0. Dotted-path strings in a Kafka connection's extra field are resolved into Python callables via import_string with no allowlist, then invoked by the confluent-kafka client. This allows someone with Kafka connection configuration access to achieve remote code execution on the Airflow Scheduler; upgrading to 2.0.0 resolves it.
VU#456290: Hugging Face Transformers library writes remote code to disk prior to consent check
CVE-2026-80047: Hugging Face Transformers 4.49.0 through 5.8.1 writes attacker-controlled Python files to disk before the trust_remote_code consent check.
CERT/CC vulnerability note VU#456290 describes CVE-2026-80047 in the Hugging Face Transformers library, affecting versions 4.49.0 through 5.8.1. The library performs a remote module fetch and writes attacker-controlled Python files to the local disk before evaluating the trust_remote_code consent prompt, without user authorization. This violates the security contract enforced across other dynamic module-loading paths in the library. Transformers is a primary framework for training and inference across NLP, vision, audio, video, and multimodal machine learning systems.
CPython: [CVE-2026-87910] tarfile hardlink fallback ignores custom extraction filter rejection via None
CPython CVE-2026-87910: tarfile's hardlink fallback ignores custom extraction filters that reject entries via None, rated MEDIUM severity.
Petr Viktorin disclosed CVE-2026-87910 on the CPython security-announce list, rated MEDIUM severity. The flaw is in the tarfile module: when a custom extraction filter rejects an entry by returning None, the hardlink fallback path ignores the rejection. The posted excerpt did not include affected versions or patch details.