ClawHub
Back to skill

Security audit

OpenClaw Auto‑Updater (Safe + Scheduled + Summary)

Security checks across malware telemetry and agentic risk

Overview

This skill is a transparent auto-updater, but it can repeatedly change OpenClaw and all installed skills without per-update review.

Use this only if you are comfortable allowing scheduled jobs to update OpenClaw and all installed skills automatically. Prefer starting with dry-run or core-only mode, review the cron schedule and update scope, and know how to edit or remove the cron job before enabling unattended live updates.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly promotes unattended automatic updates for both the core application and installed skills, but it does not prominently warn that updates can change behavior, introduce incompatibilities, restart services, or fail in ways that affect availability. In this context, the risk is operational rather than directly malicious: a user may enable hands-off updates without understanding the disruption or trust implications of automatically pulling and applying changes.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The setup confirmation example normalizes unattended daily updates of OpenClaw core and all installed skills without explicitly warning that this will make automatic system changes. In an auto-updater skill, that omission is security-relevant because users may enable recurring modification of code and dependencies without appreciating the operational and supply-chain risk.

Self-Modification

High
Category
Rogue Agent
Content
# Capture new version
OPENCLAW_VERSION_AFTER=$(openclaw --version 2>/dev/null || echo "unknown")

# Update skills
log "Updating skills via ClawHub..."
SKILL_OUTPUT=$(clawdhub update --all 2>&1) || true
Confidence
95% confidence
Finding
Update skill

VirusTotal

48/48 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.