PatchSiren cyber security CVE debrief
CVE-2026-34507 OpenClaw CVE debrief
A policy bypass vulnerability in OpenClaw's QQBot admin commands allows authenticated senders to circumvent DM-only and allowFrom policy checks. The flaw permits routing of admin commands from unauthorized senders or contexts, enabling execution of restricted behavior that policy should have blocked. The vulnerability affects OpenClaw versions prior to 2026.4.29. The CVSS 4.0 vector indicates network attack vector with low attack complexity, privileged access requirements, and partial user interaction, resulting in low confidentiality and integrity impacts. The weakness is categorized as CWE-863 (Incorrect Authorization). The CVE was published to NVD on 2026-05-29 and remains under analysis. No known exploitation in the wild or ransomware campaign use has been reported.
- Vendor
- OpenClaw
- Product
- Unknown
- CVSS
- LOW 2.3
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-05-29
- Original CVE updated
- 2026-05-29
- Advisory published
- 2026-05-29
- Advisory updated
- 2026-05-29
Who should care
Organizations operating OpenClaw-based QQBot deployments with admin command functionality enabled; security teams managing bot authorization policies; developers implementing similar DM-only or allowFrom restrictions in bot frameworks.
Technical summary
The vulnerability exists in OpenClaw's QQBot implementation where admin commands fail to properly enforce DM-only and allowFrom policy restrictions. Authenticated senders can exploit this by routing commands through contexts that should be blocked by policy, effectively bypassing intended authorization controls. The flaw stems from incorrect authorization logic (CWE-863) that permits policy check skipping. Attackers with authenticated access can leverage this to execute restricted admin behaviors from unauthorized senders or non-DM contexts. The fix in version 2026.4.29 presumably adds proper validation of sender context against policy requirements before command execution.
Defensive priority
LOW
Recommended defensive actions
- Upgrade OpenClaw to version 2026.4.29 or later to remediate the policy bypass vulnerability
- Review QQBot admin command routing logic to ensure DM-only and allowFrom policy checks are enforced consistently
- Audit admin command execution logs for unauthorized sender contexts if running affected versions
- Validate that authenticated sender identity is properly verified before policy check bypass decisions
- Monitor for security advisories from OpenClaw project for additional hardening recommendations
Evidence notes
Vulnerability disclosed via GitHub Security Advisory GHSA-w4v6-g3wm-w36c and Vulncheck advisory. Vendor identification marked as low confidence requiring review; product attribution based on reference domain candidate analysis.
Official resources
2026-05-29