ClawHub
Back to skill

Security audit

Kokochat Pairing

Security checks across malware telemetry and agentic risk

Overview

This skill is a KokoChat pairing helper that uses OpenClaw's official pairing flow and a relay tunnel, with some operational side effects users should understand before running it.

Install only if you are comfortable letting this skill create KokoChat relay state, run `openclaw qr`, connect your gateway to the configured relay, and manage KokoChat relay connector processes. After pairing, inspect `openclaw devices list` and revoke the phone if its identity or scopes do not match what you intended to grant.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill advertises generating a pairing/setup code, but it also performs host-side process orchestration: stopping other relay connectors, spawning detached processes, killing PIDs, and restarting a systemd unit. Even if intended to keep the relay available during pairing, these side effects exceed the stated purpose and create unnecessary operational control over the host.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
This code has broad host-level control by invoking systemctl, pgrep, process.kill, and spawning detached connector processes. In an agent skill context, that capability is dangerous because a user asking for a pairing code would indirectly grant the skill power to disrupt local services or terminate unrelated matching processes if assumptions fail.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dangerous_exec

Shell command execution detected (child_process).

Critical
Code
suspicious.dangerous_exec
Location
generate-kokochat-code.mjs:69