# CVE-2026-100596: OpenClaw authorization bypass in MCP configuration

Source: https://junglewise.ai/threats/cve-2026-100596-openclaw-versions-before-2026-7-1-fail-to-properly-authorize-non  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-09-26. Updated: 2026-09-26  
License: CC BY 4.0 with a link back

CVE-2026-100596 · Severity: high · CVSS 8.8 · Published 2026-09-26

OpenClaw is a system that integrates with MCP (Model Context Protocol) servers for AI-powered chat functionality. An authorization flaw allows non-owner users to inject and persist malicious MCP server commands that execute with OpenClaw's system privileges. This could lead to a complete compromise of the host system, including data theft, system modification, and service disruption.

Technologies: [Openclaw](https://junglewise.ai/threats/technologies/openclaw). Vendors: [Openclaw](https://junglewise.ai/threats/vendors/openclaw).

## Quick answers

**Is CVE-2026-100596 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-100596 being exploited in the wild.

**How severe is CVE-2026-100596?** High severity, with a CVSS score of 8.8, published 26 September 2026.

**Which products and versions are affected?** Openclaw before 2026.7.1.

**Is there a fix for CVE-2026-100596?** A patch was published on 2026-07-01: version 2026.7.1 released Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-100596 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2026-100596?** Check whether you run Openclaw at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

OpenClaw fails to properly authorize non-owner users executing /mcp set and /mcp unset commands, allowing them to persist arbitrary stdio MCP commands. An authenticated non-owner external-channel sender can inject attacker-selected MCP configuration that executes with the OpenClaw process user's privileges when the MCP configuration loads. This is a missing authorization check (CWE-862) combined with OS command injection (CWE-78) that affects confidentiality, integrity, and availability.

## Affected products

- OpenClaw OpenClaw before 2026.7.1

## Timeline

- 2026-09-26: disclosed
- 2026-07-01: patched: version 2026.7.1 released

## References

- [https://github.com/openclaw/openclaw/security/advisories/GHSA-wwx7-573h-pqwc](https://github.com/openclaw/openclaw/security/advisories/GHSA-wwx7-573h-pqwc)
- [https://www.vulncheck.com/advisories/openclaw-before-2026.7.1-authorization-bypass-via-mcp-configuration](https://www.vulncheck.com/advisories/openclaw-before-2026.7.1-authorization-bypass-via-mcp-configuration)

## Related threats

- [OpenClaw incorrect authorization in mcp.app.view method](https://junglewise.ai/threats/cve-2026-102807-openclaw-before-2026-9-4-contains-an-incorrect-authorization)
- [OpenClaw incorrect authorization in Gateway media root allowlist](https://junglewise.ai/threats/cve-2026-102806-openclaw-before-2026-9-5-contains-an-incorrect-authorization)
- [OpenClaw exec approval bypass in Google Meet node commands](https://junglewise.ai/threats/cve-2026-100599-openclaw-versions-2026-5-1-through-2026-7-0-fail-to-apply-the)
- [OpenClaw Signal approval reaction binding logic error](https://junglewise.ai/threats/cve-2026-100598-openclaw-npm-package-openclaw-before-2026-7-1-incorrectly-binds)
- [OpenClaw time-of-check time-of-use race condition in filesystem operations](https://junglewise.ai/threats/cve-2026-100597-openclaw-npm-package-openclaw-before-2026-7-1-is-vulnerable-to-a)