/

CVE-2026-35868

Severity:

CRITICAL RISK

(9.8)

Command Injection (CWE-20) in LB-link AC2100 Router

CVE-2026-35868

Deferred

Vulnerability type

Category:

Command Injection

CWE:

CWE-20

Impact:

Arbitrary OS Command Execution and Router Compromise

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 command injection vulnerability in the LB-link Router AC2100 allows attacker-controlled input to reach an operating system command execution context through the bs_SetLimitCli_info function.

If the vulnerable parameter is reachable and accepts insufficiently validated input, an attacker may execute arbitrary commands with the privileges of the affected router process.

Affected context

The vulnerability occurs in the libshare.so library included with LB-link Router AC2100_AZ3 V1.0.4, specifically within the bs_SetLimitCli_info function. Risk depends on whether an attacker can reach the relevant management or service interface and provide input that is later incorporated into a system-level command without adequate validation or separation from command syntax.

Exploit preconditions

  • Attack surface: A router service, management interface, or request path that invokes bs_SetLimitCli_info
  • Required access: Ability to submit or influence the vulnerable parameter; authentication requirements depend on the deployment and interface exposure
  • Trigger point: Processing of the attacker-controlled value by the affected function
  • Why exploitation is possible: Input is insufficiently constrained before being passed into a command execution context, allowing shell control characters or command syntax to alter execution behavior

Why this happens

The underlying issue is improper input validation and neutralization of data used in an operating system command context.

When application data is concatenated into a shell command or otherwise interpreted as command syntax, untrusted characters can change the meaning of the intended operation. This behavior maps to CWE-20: Improper Input Validation, with command injection occurring as the resulting security impact.

Potential impact

  • An attacker may execute arbitrary operating system commands on the router
  • The attacker could modify configuration, services, or stored data
  • The device may be used to monitor, redirect, or disrupt network traffic
  • Compromise could enable persistence or lateral movement within connected networks
  • Router availability and integrity could be impaired

General mitigation guidance

Mitigation generally focuses on removing shell interpretation from application workflows and reducing trust in externally supplied values.

Common defensive patterns include:

  • Replacing shell command construction with dedicated APIs that pass arguments separately
  • Applying allowlist validation for expected values, formats, ranges, and lengths
  • Treating user-controlled data strictly as data rather than executable command syntax
  • Avoiding shell invocation where an equivalent library or system interface is available
  • Running network-facing services with the minimum privileges required
  • Restricting management interfaces and separating administrative functions from untrusted networks
  • Using defense-in-depth controls such as process isolation and filesystem restrictions where supported

How teams detect issues like this

Organizations can identify command injection risks through:

  • Static analysis rules for shell invocation, command concatenation, and unsanitized arguments
  • Code review checks for system calls receiving request, configuration, or form-derived values
  • Firmware and binary analysis focused on command execution paths in shared libraries
  • Runtime monitoring for unexpected child processes, shell launches, or unusual command-line arguments
  • Policy enforcement requiring allowlist validation and non-shell APIs for system operations
  • Automated security review tools that flag untrusted data flows into command execution sinks

Sources

SourceLink
github.comgithub.com/Orcust-Automaton/Vulnerability/blob/main/LB-Link/AC2100_AZ3/bs_SetLimitCli_info.md

Frequently asked

questions

It may be remotely exploitable if the affected function is reachable through a network-accessible router service and an attacker can influence its vulnerable parameter. Actual exposure depends on interface availability, authentication, and deployment configuration.

The attacker needs a path to submit or modify input consumed by bs_SetLimitCli_info. Whether that requires authentication depends on the exposed service and router configuration.

The safest pattern is to avoid shell interpretation entirely, use structured APIs with separate arguments, and enforce strict allowlist validation before processing input.

Minimize command execution, review all data flows into system-level APIs, apply least privilege, restrict management interfaces, and include command-injection checks in firmware testing and code review.

Find out what's already
exploitable in your codebase.