ClawHub
Back to skill

Security audit

Workflow Catalyst Free

Security checks for vulnerabilities and agentic risk

Overview

This skill is an automation helper, but it pushes agents toward proactive script creation and execution with broad triggers and limited consent guidance.

Review this skill carefully before installing. It is not showing malicious payloads, but it may cause an agent to propose or create automations too aggressively. Only use it where you expect script generation, and require the agent to ask before reading private files, using credentials, sending email, calling APIs, modifying systems, or setting up scheduled/background tasks.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Vague Triggers

High
Confidence
90% confidence
Finding
The activation criteria are overly broad and encourage the agent to invoke automation behavior from common conversational cues like 'I have to do this again' or mention of copying/exporting. In a skill with exec capability, this raises the chance of unsolicited script generation or workflow execution without sufficiently explicit, informed user intent.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The keyword-based capability coverage is vague and expansive, making it easier for the skill to be invoked in loosely related contexts without firm boundaries. In combination with automation-oriented behavior and exec access, ambiguous scope increases the risk of overreach into tasks the user did not intend to automate.

Missing User Warnings

High
Confidence
95% confidence
Finding
The skill describes automated file monitoring, data collection, email sending, and API-based synchronization without user-facing warnings about privacy, credential handling, data exfiltration, or system-side effects. These operations can touch sensitive files, send data externally, or alter systems, so omitting safety guidance materially increases the chance of harmful misuse.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill explicitly encourages the agent to 'directly do it' rather than wait for a user request, which promotes autonomous script creation and possible execution without adequate authorization or review. In a skill allowed to use exec, this can lead to unintended system changes, network actions, scheduled jobs, or handling of sensitive data based only on inferred repetition.

Static analysis

No suspicious patterns detected.