ClawHub
Back to skill

Security audit

Coze 工作流执行 / Coze Workflow

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Coze workflow caller that needs a Coze API key and sends workflow inputs to Coze, with no hidden code or automatic persistence found.

Install this only if you want your agent to call Coze workflows. Treat the Coze API key as a secret, protect the config file, avoid sending personal data or credentials as workflow parameters, and require confirmation or allowlists for workflows that spend credits, publish content, or modify external data.

SkillSpector

By NVIDIA
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 (1)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The documentation includes a plaintext API key field and example credential placement in a user config path, but provides no warning about secret handling, file permissions, redaction, or avoiding commits/logging. In a base skill that other integrations depend on, this increases the chance that users will store, copy, or expose long-lived credentials insecurely across multiple downstream skills.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.