ClawHub
Back to skill

Security audit

server-health

Security checks for vulnerabilities and agentic risk

Overview

This skill is a read-only server health checker whose sensitive system details are disclosed and aligned with its monitoring purpose.

Install only if you want a local read-only diagnostic tool. Its output can reveal operational details such as process names, gateway port, model choices, and OpenClaw session counts, so only run or forward it in trusted admin contexts.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The script reads detailed OpenClaw configuration from /root/.openclaw/openclaw.json, including gateway port, primary model, and fallback model chain, which goes beyond basic health monitoring. In a Telegram/CLI health-check context, exposing root-owned configuration data can leak operational details to less-privileged users or remote chat recipients, aiding reconnaissance and revealing sensitive deployment metadata.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The script accesses files under /root without warning the user that sensitive configuration and session-related data may be inspected. Because this skill is intended for Telegram or CLI health checks, the lack of disclosure and consent increases the chance that privileged data will be surfaced in a chat or other shared interface where it does not belong.

Context Leakage

High
Category
Data Exfiltration
Content
# Get from session status - parse the output
    local status=$(openclaw status 2>/dev/null || echo "")
    
    # Try to extract context and tokens
    local context=$(echo "$status" | grep -oP 'Context:.*?\K\d+k/\d+k \(\d+%\)' | head -1 || echo "N/A")
    local tokens=$(echo "$status" | grep -oP 'Tokens:.*?\K\d+ in / \d+ out' | head -1 || echo "N/A")
Confidence
91% confidence
Finding
The code is designed to parse model context and token usage from openclaw status, which can reveal session activity characteristics and potentially sensitive usage metadata. In the stated Telegram/CLI monitoring use case, even aggregate context/token information may leak details about active agent sessions to unauthorized viewers and expands the script from health monitoring into session introspection.

Static analysis

No suspicious patterns detected.