ClawHub
Back to skill

Security audit

GREEN-API

Security checks for vulnerabilities and agentic risk

Overview

This WhatsApp integration is purpose-aligned, but it gives an agent broad access to messages, contacts, group administration, and account controls without enough safety guidance.

Install only if you trust this skill and the GREEN-API MCP endpoint with your WhatsApp account. Use least-privilege or non-production credentials where possible, avoid pasting real API tokens into ordinary chat, limit history reads to specific chats and small counts, and require explicit confirmation before sending, deleting, changing settings, logging out, rebooting, modifying groups, or deleting instances.

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 (4)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The README explicitly instructs the user to paste a live access token into a natural-language chat message. Chat prompts may be logged in client history, telemetry, screenshots, transcripts, or passed through intermediary services, so presenting credentials this way encourages unsafe secret handling and increases the chance of credential disclosure.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly instructs users to provide `instance_id` and `api_token` and says the credentials are stored for the session, but it does not clearly warn about the sensitivity of those secrets or advise users not to expose, reuse, or share them. In an agent setting, omission of that warning can lead users to provide production credentials without understanding the trust and retention implications, increasing the chance of account compromise or unintended message access.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill exposes tools to read chat history, fetch specific messages, and retrieve recent incoming/outgoing messages, but it does not warn that these actions may reveal private conversations, personal data, or sensitive business communications. In a WhatsApp context, this is especially risky because users may not realize the agent can access broad historical message content across contacts and groups.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill includes destructive or high-impact actions such as deleting messages, logging out, rebooting instances, altering settings, and removing group participants, yet it does not warn users that these actions may be irreversible or service-disruptive. In this context, an agent could carry out destructive account or messaging changes under ambiguous instructions, causing loss of data, interruption of service, or accidental administrative changes.

Static analysis

No suspicious patterns detected.