ClawHub
Back to skill

Security audit

Connect Mermail Composio

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Mermail Composio integration guide that can affect connected third-party accounts, but its authority is purpose-aligned and includes user-auth and safety checks.

Install only if you intend to let Mermail connect to and operate on selected third-party services. Review each toolkit's permissions during OAuth or hosted auth, and treat executions as actions in those external accounts, even though the skill includes connection checks and destructive-action safeguards.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The description says to use the skill whenever a user needs apps like GitHub, Slack, Notion, or Google Calendar for a Mermail workflow, or when checking whether a toolkit is connected. This is a wide, open-ended activation condition without clear exclusions or negative examples, which could cause unintended invocation for many ordinary requests involving those services.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The markdown explains how to connect and execute Composio tools against third-party apps, but it does not explicitly warn users that their data or instructions may be sent to external services and that actions may affect third-party accounts. Because the skill covers services like GitHub, Slack, Apollo, Notion, and Google Calendar, a clear disclosure about privacy and account-impacting behavior is warranted.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.