ClawHub
Back to skill

Security audit

Flowforge Builder Free

Security checks for vulnerabilities and agentic risk

Overview

This is a workflow-automation skill whose file, network, command, watch, and cron examples are disclosed and aligned with its stated purpose, though users should scope workflows carefully.

Install only if you intend to let the agent help define or run automation workflows. Review each workflow before execution, keep watched directories narrow, avoid sending sensitive file contents to external APIs, and treat shell commands and cron entries as privileged actions.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly exposes high-impact capabilities—file writes, file moves, shell command execution, and outbound network requests—without any safety warnings, confirmation requirements, path restrictions, or trust-boundary guidance. In an agent setting, this can lead users to run workflows that modify local files, exfiltrate data, or execute arbitrary commands without understanding the risk.

External Transmission

Medium
Category
Data Exfiltration
Content
{ "action": "read", "file": "${trigger.file}", "output": "fileContent" },

    // 第二步：用fileContent作为输入
    { "action": "fetch", "url": "https://api.example.com/process", "method": "POST", "body": "${fileContent}", "output": "result" },

    // 第三步：用result作为输入
    { "action": "save", "path": "./output/result.json", "input": "${result}" }
Confidence
89% confidence
Finding
This example posts `${fileContent}` from a watched local file directly to an external URL, demonstrating a pattern that can exfiltrate newly created local data off-host. In the context of an agent skill that also supports file monitoring, this materially increases danger because users may adopt the example without recognizing that local file contents are being transmitted externally.

Static analysis

No suspicious patterns detected.