Vulnerability details

CVE-2026-27952

Agenta-AI agenta-api

Agenta has Python Sandbox Escape, Leading to Remote Code Execution (RCE)

CVSS 3.1
8.8High

Official description

Agenta is an open-source LLMOps platform. In Agenta-API prior to version 0.48.1, a Python sandbox escape vulnerability existed in Agenta's custom code evaluator. Agenta used RestrictedPython as a sandboxing mechanism for user-supplied evaluator code, but incorrectly whitelisted the `numpy` package as safe within the sandbox. This allowed authenticated users to bypass the sandbox and achieve arbitrary code execution on the API server. The escape path was through `numpy.ma.core.inspect`, which exposes Python's introspection utilities — including `sys.modules` — thereby providing access to unfiltered system-level functionality like `os.system`. This vulnerability affects the Agenta self-hosted platform (API server), not the SDK when used as a standalone Python library. The custom code evaluator runs server-side within the API process. The issue is fixed in v0.48.1 by removing `numpy` from the sandbox allowlist. In later versions (v0.60+), the RestrictedPython sandbox was removed entirely and replaced with a different execution model.

CVSS scores

ScoreVersionSourceVector
8.8High 3.1 CVEProject cvelistV5 (CNA) (GitHub_M) CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Affected products

VendorProductVersionsSource
Agenta-AI agenta-api < 0.48.1 CVEProject cvelistV5 (CNA)

References and advisories

1
Kolun on Telegram @kolunsec Exploited CVEs, key news and a daily digest Alerts for this product's next CVEsFollow the product to get new vulnerabilities on Telegram.

Details

Published
2026-02-26
Last updated
2026-02-26
State
Published
Assigner (CNA)
GitHub_M

Exploit prediction (EPSS)

0.56%
Probability of exploitation in 30 days
47.1%
Percentile (among all CVEs)

SSVC assessment (CISA)

Automatable
No
Exploitation
Proof of concept
Technical impact
Total

Weaknesses (CWE)

  • CWE-94
    Improper Control of Generation of Code ('Code Injection')