PT-2026-53481 · Pypi · Llama-Index-Core

Published 2026-06-29 · Updated 2026-06-29

OSV

CVSS v3.1

9.8

Critical

Base vector

AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

A command injection vulnerability exists in the run-llama/llama index repository, specifically within the safe eval function. Attackers can bypass the intended security mechanism, which checks for the presence of underscores in code generated by LLM, to execute arbitrary code. This is achieved by crafting input that does not contain an underscore but still results in the execution of OS commands. The vulnerability allows for remote code execution (RCE) on the server hosting the application.

Fix

Related Identifiers

PYSEC-2026-396

Affected Products

Llama-Index-Core

References · 8

• https://nvd.nist.gov/vuln/detail/CVE-2024-3271 · Security Note
• https://osv.dev/vulnerability/PYSEC-2026-396 · Vendor Advisory
• https://github.com/run-llama/llama_index/commit/2c92e88838a5f481d50840240b1dd3180066c6f5 · Note
• https://github.com/run-llama/llama_index/commit/5fbcb5a8b9f20f81b791c7fc8849e352613ab475 · Note
• https://github.com/run-llama/llama_index · Note
• https://pypi.org/project/llama-index-core · Note
• https://huntr.com/bounties/9b32490e-7cf9-470e-8d49-ba083ae7a279 · Note
• https://github.com/advisories/GHSA-r6gp-rff2-p3hf · Note