SEPTEMBER 19, 2026
Live Feed
Back to database
Case File

CVE-2026-62677

HIGH · CVSS 8.8 EPSS 0.45% Public Exploit

Source: NVD + CISA KEV + EPSS · Published 2026-08-21 · Last synced 2026-09-18

CyberRota Analysis

AI-Generated

Authenticated users of the Omnigent AI agent framework prior to version 0.3.0 can exploit a vulnerability that allows them to upload a session-scoped agent bundle with a manipulated `os_env.cwd` value, leading to unauthorized access to files and environment secrets outside the designated workspace. This flaw poses a significant risk as it enables attackers to execute system-level read, write, and shell commands, compromising the integrity and confidentiality of the runner's environment. Organizations utilizing Omnigent should prioritize upgrading to version 0.3.0 to mitigate this high-severity vulnerability.

Public Exploit Signal

A public exploit, PoC, GitHub repository or Metasploit reference was detected for this CVE.

Note: these links are listed for security research and verification purposes only.

CVE
CVE-2026-62677
Severity
HIGH
CVSS
8.8
EPSS
0.45%

Original NVD Description

Omnigent is an open-source AI agent framework and meta-harness for orchestrating coding agents. Prior to 0.3.0, an authenticated user can upload a session-scoped agent bundle with an absolute or traversal-containing os_env.cwd value because omnigent/spec/parser.py stores the value verbatim and omnigent/spec/validator.py does not constrain it. On a runner where OMNIGENT_RUNNER_WORKSPACE is unset, omnigent/runner/resource_registry.py preserves the attacker-controlled path and omnigent/inner/os_env.py uses the resolved path as the environment root and copytree source. The _assert_within_cwd check then treats that attacker-selected root as trusted, allowing sys_os_read, write, edit, and shell tools to access runner files and environment secrets outside the intended workspace. This issue is fixed in version 0.3.0.