ClawHub

Email Outreach

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a disclosed email outreach helper with local templates/scripts and no artifact-backed evidence of hidden or unrelated behavior.

Install only if you want an agent to help with outreach workflows. Review bundled scripts and reference files before first use, keep lead/contact data scoped to the intended project, and require explicit approval before any real email sending or automation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill instructs the agent to read local files such as `references/copywriting-frameworks.md`, `assets/sequence-template.md`, and to run `scripts/deliverability_check.py`, but no permissions are declared. That creates an undeclared capability boundary: users and platform policy may assume a purely textual skill while it actually depends on file access and script execution paths, which can expand attack surface and enable unintended data access or tool use.

Tp4

High
Category
MCP Tool Poisoning
Confidence
81% confidence
Finding
The declared description promises broad campaign generation, personalization, subject-line generation, and lead-list building, while the implemented behavior appears much narrower and includes references to local resources/scripts instead. This mismatch is dangerous because it misleads users and policy systems about what the skill actually does, weakening trust boundaries and potentially causing the skill to be invoked in contexts where its real behavior is not appropriate or has not been reviewed.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The trigger phrases are very broad, including common terms like "email templates," "sales emails," and "lead outreach," without clear activation boundaries. Overbroad matching can cause the skill to activate unexpectedly on ordinary user requests, increasing the chance of unintended file access, use of internal references, or application of the wrong workflow in a sensitive context.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal