ClawHub
Back to skill

Security audit

HeyLead

Security checks across malware telemetry and agentic risk

Overview

HeyLead appears purpose-built for LinkedIn sales automation, but it needs Review because it can act from a user’s LinkedIn account autonomously and its approval, token, and scheduler boundaries are not clear enough.

Install only if you are comfortable letting a third-party MCP package operate your LinkedIn account. Pin and inspect the `heylead` package before use, start in Copilot/review mode, keep the cloud scheduler disabled until safeguards are clear, verify how to revoke Google/LinkedIn access, and avoid using it for sensitive conversations unless you accept the local storage and external AI processing model.

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 (3)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The example utterances are short, generic natural-language phrases like "Check my replies" and "Suggest next action" that could plausibly appear in ordinary conversation and unintentionally trigger the skill. In this skill's context, accidental activation is more dangerous because the toolset can perform real external actions on LinkedIn, including sending outreach and publishing content.

Missing User Warnings

High
Confidence
97% confidence
Finding
The skill advertises autonomous outreach, follow-ups, reply handling, scheduling, and content publishing without a prominent warning that it can take external actions on the user's LinkedIn account. That omission raises the risk that users or upstream agents enable automation without understanding that messages, engagements, and posts may be sent to third parties automatically.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The manifest explicitly promotes fully autonomous LinkedIn outreach, automatic follow-ups, and autonomous reply handling, but provides no warning that the skill can send external messages, impersonate the user at scale, or trigger account, compliance, and reputation consequences. In this context, the absence of clear consent boundaries and safety disclosures increases the risk of unintended spam, policy violations, and unauthorized business communications.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.