CVE-2026-100552

8.8 HIGHpublic exploit available

Published 2026-09-26 · Updated 2026-09-30

AI risk analysis

Summary: The flaw in OpenClaw allows bypassing of tool restrictions in Codex app-server, enabling access to native command and file tools even when a tool allowlist is in place.
Exploitability: Exploitation requires access to trigger the Codex runtime agent in a lower-trust conversation, making it moderately exploitable.
Blast radius: If exploited, the impact could be significant, allowing unauthorized access to system-level commands and files, potentially leading to full system compromise.
Detection: No reliable host or network indicator is derivable from the published description.
Prioritized remediation: Upgrade to OpenClaw version 2026.8.1 or later.

rceauth-bypassweb

Analysis generated locally by qwen2.5:7b-instruct (no data left the box). AI-assisted — verify against primary sources before acting.

NVD description

OpenClaw (npm package 'openclaw') before 2026.8.1 does not correctly enforce per-chat tool policies for Codex app-server runtime tools. A conversation-level tools.allow rule filtered OpenClaw tools but did not restrict the shell, process, file, and patch tools owned by the Codex runtime. When a lower-trust conversation was assigned to a Codex runtime and restricted with a per-chat tool allowlist, a participant able to trigger that agent could still reach native command and file tools, bypassing the configured allowlist. The practical impact depends on the runtime's host permissions and sandbox configuration. The issue is fixed in 2026.8.1.

CVSS vector

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Weaknesses

CWE-863

Public exploit & PoC references

https://github.com/openclaw/openclaw/security/advisories/GHSA-wwcw-jfpp-gpxw

All references

Source data: NVD (nvd.nist.gov), public domain. Exploit-DB.ai adds local AI analysis for defensive use only.

Related CVEs

Related by shared AI tags and CWE weakness class. Browse the full archive.