ClawHub
Back to skill

Security audit

OpenClaw Tmux Persistent Process

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed tmux helper for keeping user-chosen long-running processes alive, with some overbroad wording but no hidden or destructive behavior found.

Install this only if you want the agent to use tmux for explicitly intended long-running tasks. Review active sessions and clean them up when finished, especially for tunnels, background workers, or commands that inherit sensitive environment variables.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · SKILL.md (reported line 21)May include surrounding context.

md
- **Background processes are tied to the exec session.** When the session cleans up (idle timeout, context compaction, or gateway restart), all background processes receive SIGKILL and die immediately.
- **Gateway restarts kill everything.** Updates, config changes, or `openclaw gateway restart` — all running exec processes are gone.
- **No warning.** Your tunnel, dev server, or build task just silently disappears.

This means anything you expect to keep running — a tunnel for a client demo, a dev server, a long build — can die at any moment without notice.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The description recommends the skill for broadly defined long-running processes and even says it is 'recommended for all OpenClaw users,' which can cause the agent to over-activate this skill outside situations where persistence is actually needed. Overly broad trigger language increases the chance of unnecessary use of persistent background execution, which can expand operational risk and reduce user control over long-lived processes.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The heading 'Why every OpenClaw user should have this' acts as a broad activation cue and frames the skill as universally applicable rather than conditionally useful. In an agent setting, this can bias selection toward persistent-process behavior even when a simpler or safer non-persistent command would be more appropriate.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.