OpenClaw Chain of Execution (CoE)

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:25 PM UTC
• Version:
  1.0.0

Overview

This skill is a lightweight progress-narration helper; its broad triggers may make responses more verbose, but it does not access data or perform risky actions.

Install this only if you want more visible step-by-step progress updates during tasks. Be aware that generic trigger words may make it activate unexpectedly and produce more process detail than usual; disable it if that becomes noisy or reveals more workflow detail than you want.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The trigger list includes very generic terms such as “过程” and “一步一步”, which are common in ordinary conversation and can cause the skill to activate unintentionally. In a skill that changes output behavior to reveal step-by-step execution details, accidental activation can expose internal workflow information or degrade the expected interaction model.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The automatic activation rule is underspecified because it says the skill enables itself when users include vague keywords like '分步' or '执行过程' without defining strict matching or consent requirements. This increases the chance of unintended activation and, given the skill’s purpose, may lead to oversharing of internal execution details that should not be automatically surfaced.

VirusTotal

65/65 vendors flagged this skill as clean.