CVE-2026-53806: OpenClaw: Combined POSIX shell options could confuse exec revalidation
OpenClaw's exec revalidation can read combined POSIX shell options differently at approval time and at execution time. A command using combined shell flags can then run inline shell content the allowlist never approved. Affects openclaw up to 2026.5.7. The first stable patched version is 2026.5.12.
Published Jul 2, 2026 · Analysis Oct 25, 2026
What an attacker can do
OpenClaw checks a command against an allowlist, then revalidates it before running. With combined short flags, the two parses of the shell options disagree. The weakness class is time-of-check versus time-of-use (CWE-367). The outcome is that inline shell content, the script passed to the shell, runs without the allowlist decision that should have covered it.
The advisory names no specific flag combination and the sources give no more detail on the parser difference. It says practical impact depends on the operator's configuration and on whether lower-trust input can reach the exec path. It also keeps OpenClaw's trusted-operator model: authenticated Gateway operators and installed plugins stay trusted.
The risk is a Gateway with an exec allowlist, such as a permitted shell, where an agent or a channel user lower in trust can submit command requests. A prompt-injected agent could use the flaw to run arbitrary shell content through an allowlisted shell. The score is 8.8 under CVSS 3.1: network, low privileges, no interaction.
Who is affected
Affected: npm package openclaw 2026.5.7 and earlier, where the exec feature with an allowlist is enabled and reachable by users or agents who are not fully trusted.
npm ls openclawReview your exec allowlist for shell entries that accept options such as combined flag forms.
openclaw
- Ecosystem
- npm
- Affected
- <= 2026.5.7
- Fixed in
- 2026.5.12
- Check
npm ls openclaw
How to fix it
Upgrade to openclaw 2026.5.12 or later, the first stable patched version.
npm install -g [email protected]Until then, the advisory says to avoid combined shell option forms in allowlisted commands. It also advises keeping channel and tool allowlists narrow, not sharing one Gateway between mutually untrusted users, and disabling exec when it is not needed.
How to tell if you were hit
Review exec logs for allowlisted shell commands that use combined option forms, and compare the approved command text with what actually ran. Look for shell processes started by the Gateway with inline script arguments you did not expect. The advisory publishes no log strings, so this is a review of your own records.
Original advisory
From the GitHub Advisory Database (CC BY 4.0).
Summary
Combined POSIX shell options could confuse exec revalidation. In affected versions, a command request using combined shell flags could parse approval-time and execution-time shell options differently.
This advisory is scoped to the named feature and configuration. It does not change OpenClaw's trusted-operator model: authenticated Gateway operators, installed plugins, and intentional local execution surfaces remain trusted unless a separate policy, approval, allowlist, sandbox, or auth boundary is crossed.
Impact
When the affected feature is enabled and reachable, this could run inline shell content without the intended allowlist decision. Practical impact depends on the operator's configuration and whether lower-trust input can reach that path.
Patched Versions
The first stable patched version is 2026.5.12.
Mitigations
avoid combined shell option forms in allowlisted commands until patched. As general hardening, keep channel and tool allowlists narrow, avoid sharing one Gateway between mutually untrusted users, and disable the affected feature when it is not needed.
Credited to YLChen-007.
Questions
- Which shell flags trigger CVE-2026-53806?
The sources only say combined POSIX shell flags. They do not name specific combinations, so treat any combined short-option form in an allowlisted shell command as suspect.
- Does this break OpenClaw's trust model?
The advisory says it does not change the trusted-operator model. Authenticated Gateway operators and installed plugins stay trusted. The flaw matters when lower-trust input can reach the exec path.
- Which version fixes it?
2026.5.12 is the first stable patched version. The package data lists 2026.5.7 and earlier as vulnerable.
References
How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. The analysis is written by an AI model from the sources listed above and checked by automated rules before it is published. Check versions against the vendor's advisory before you act. Methodology
More OpenClaw vulnerabilities
All OpenClaw entriesCVE-2026-62197
OpenClaw: Browser CDP discovery could accept blocked WebSocket URLs
medium8.5Oct 21, 2026CVE-2026-35630
OpenClaw: QQBot native approval buttons did not enforce configured approver identity
high8.0Jul 2, 2026CVE-2026-53809
OpenClaw: Embedded runner policy could be confused by provider aliases
medium3.8Jul 2, 2026CVE-2026-53810
OpenClaw's marketplace runtime extension metadata could point at unscanned payloads
high8.8Jul 2, 2026CVE-2026-53811
OpenClaw: Matrix allowFrom could bind to mutable display names
high8.8Jul 2, 2026CVE-2026-53812
OpenClaw's browser act interactions could bypass private-network navigation checks
medium7.7Jul 2, 2026
