ClawHub
Back to skill

Security audit

Gupshup

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Gupshup connector wrapper that can send messages, but it clearly labels write actions and requires user confirmation before changing state.

Install only if you intend to let the agent operate your connected Gupshup account. Review any proposed send_template_message or send_text_message payload carefully, because those actions can send real WhatsApp messages through your account.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger description is intentionally broad: 'Use this skill for ANY Gupshup request' and 'Whenever a task involves Gupshup, use this skill instead of calling the API directly.' This can cause the agent to route all Gupshup-related tasks through a write-capable skill, increasing the chance of unintended invocation and accidental state-changing actions if task classification or confirmation logic is weak.

Static analysis

No suspicious patterns detected.