/

CVE-2026-85425

Severity:

CRITICAL RISK

(9.8)

OS Command Injection (CWE-78) in MOOS-IvP iSay Handler

CVE-2026-85425

Analyzed

Vulnerability type

Category:

OS Command Injection

CWE:

CWE-78

Impact:

Remote Code Execution

Severity:

CRITICAL RISK

(9.8)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

A critical OS command injection vulnerability in the MOOS-IvP iSay message handler allows shell metacharacters in published SAY_MOOS messages to be interpreted as commands. An attacker who can publish messages may execute arbitrary commands with the privileges of the iSay process.

Affected context

The vulnerability occurs in MOOS-IvP iSay deployments that process SAY_MOOS messages through the affected variable handler and pass message content to a shell command without adequate separation between data and executable arguments. Risk depends on whether an attacker can publish messages to the relevant MOOS communication channel and on the operating-system privileges granted to the iSay process.

Exploit preconditions

  • Attack surface: The MOOS messaging interface handling SAY_MOOS values
  • Required access: Ability to publish or inject messages accepted by the iSay process
  • Trigger point: Processing of a SAY_MOOS message by the vulnerable handler
  • Why exploitation is possible: Unsanitized message content is incorporated into a shell command, allowing shell syntax to alter command execution

Why this happens

The underlying bug is improper neutralization of special elements used in an operating-system command. The handler treats externally influenced message text as part of a shell command rather than as data passed through a safe argument interface. This behavior maps to CWE-78: Improper Neutralization of Special Elements used in an OS Command, commonly known as OS command injection.

Potential impact

  • Arbitrary commands may execute as the iSay process user
  • The host or vehicle-control environment could be compromised
  • Sensitive files or operational data may be accessed or modified
  • Attackers could disrupt services or alter mission-related behavior
  • Compromise could enable lateral movement where additional privileges or network access are available

General mitigation guidance

Mitigation generally focuses on eliminating shell interpretation for message-derived data and reducing the trust placed in messaging inputs.

Common defensive patterns include:

  • Invoking programs directly with structured argument arrays instead of shell command strings
  • Applying strict allowlists for supported speech or action values
  • Separating message parsing, validation, and command execution
  • Rejecting unexpected control characters and shell metacharacters where appropriate
  • Running the iSay process with the minimum required operating-system privileges
  • Restricting which components and network identities can publish relevant MOOS messages
  • Applying filesystem, process, and network containment appropriate to the deployment

How teams detect issues like this

Organizations can look for command-injection risks through:

  • Static analysis rules identifying shell execution APIs that receive message, network, or configuration data
  • Code review checks for string concatenation or interpolation used to construct commands
  • Tests that verify message values are passed as literal arguments rather than interpreted shell syntax
  • Runtime monitoring for unexpected child processes spawned by the iSay service
  • Message-bus authorization and policy enforcement that limits publishers of SAY_MOOS data
  • Automated security review tools that flag unsanitized data flows into process-launching functions

Sources

SourceLink
github.comgithub.com/moos-ivp/moos-ivp
github.comgithub.com/moos-ivp/moos-ivp/blob/1de9ae146cd63c209e8c3fd81611a4ed2472971b/ivp/src/iSay/Sayer.cpp#L416
github.comgithub.com/moos-ivp/moos-ivp/commit/ed3a44131fd3e5528602f20adc37ba82273c1cd8
github.comgithub.com/moos-ivp/moos-ivp/pull/120
vulncheck.comvulncheck.com/advisories/moos-ivp-through-24.8.1-isay-command-injection-via-say-moos

Frequently asked

questions

It may be exploitable remotely when the MOOS messaging infrastructure is reachable by an attacker or by a compromised component that can publish accepted SAY_MOOS messages.

The attacker generally needs the ability to publish messages to the relevant MOOS channel. The practical scope depends on message-bus authentication, authorization, network segmentation, and deployment configuration.

Avoid passing message content through a shell. Use direct process APIs with fixed executables and separately validated arguments, while allowing only the values required by the application.

Enforce publisher authorization, validate message values against explicit schemas or allowlists, review all process-launching code, and run services with minimal privileges and appropriate containment.

Find out what's already
exploitable in your codebase.