Maliciously Crafted Model Archive Can Lead To Arbitrary File Write

CAO 7.1CVE-2021-41127 · GHSA-4365-fhm5-qcrx · PYSEC-2021-381 · Công bố 21/10/2021 · cập nhật 17/06/2026

Tóm tắt nhanh

CVE-2021-41127 là lỗ hổng mức CAO (CVSS 7.1) ảnh hưởng rasa, rasa (rasa).
Xác suất bị khai thác trong 30 ngày tới (EPSS): 0.76%, cao hơn 55.4% các lỗ hổng khác.
Cách xử lý: cập nhật lên 2.8.10 hoặc mới hơn.

Gói / sản phẩm	Phiên bản bị ảnh hưởng	Phiên bản đã vá
rasa PyPI	`< 2.8.10`	`2.8.10`
rasa rasa	`< 2.8.10`	`2.8.10`

Mô tả

Impact

An Archive Extraction (Zip Slip) vulnerability in the functionality that allows a user to load a trained model archive in Rasa 2.8.9 and older allows an attacker arbitrary write capability within specific directories using a malicious crafted archive file.

Patches

The vulnerability is fixed in Rasa 2.8.10

Workarounds

Mitigating steps for vulnerable end users are to ensure that they do not upload untrusted model files, and restrict CLI or API endpoint access where a malicious actor could target a deployed Rasa instance.

For more information

If you have any questions or comments about this advisory:

Email the Rasa Security Team

Mô tả của NVD

Rasa is an open source machine learning framework to automate text-and voice-based conversations. In affected versions a vulnerability exists in the functionality that loads a trained model `tar.gz` file which allows a malicious actor to craft a `model.tar.gz` file which can overwrite or replace bot files in the bot directory. The vulnerability is fixed in Rasa 2.8.10. For users unable to update ensure that users do not upload untrusted model files, and restrict CLI or API endpoint access where a malicious actor could target a deployed Rasa instance.