ClawHub
Back to skill

Security audit

producthunt-wingman

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent about being an outreach automator, but it asks for account-impacting LinkedIn automation through missing server code, broad triggers, scheduling, and a saved browser session.

Review before installing. Do not let it run setup.sh, main.py, or LinkedIn automation until the referenced server files are available from a trusted source and have been inspected. Use a dedicated LinkedIn browser profile, confirm exactly what messages or connection requests can be sent automatically, verify how the cron schedule is installed and removed, and know how to delete the saved browser profile and prospect data.

SkillSpector

By NVIDIA
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 (4)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger phrase "Start my PH outreach" is broad enough that a normal user request about beginning outreach could unintentionally invoke this skill. In this skill's context, invocation is more dangerous because the skill can start a local server, launch automated pipelines, and drive LinkedIn outreach actions, so accidental activation could cause unintended network activity and account-impacting automation.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrase at line 19 ("Start my PH outreach") is broad enough that an ordinary user request about beginning outreach could unintentionally activate this skill. Because the skill automates outreach and engagement actions on a schedule, accidental invocation could cause unintended scraping, enrichment, or messaging activity.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrase at line 23 ("Enrich new prospects") is too generic and could match common CRM or sales-assistant requests unrelated to this skill. In the context of a tool that performs prospect enrichment and LinkedIn engagement, ambiguous activation increases the risk of unintended processing of contacts or external actions.

Vague Triggers

High
Confidence
98% confidence
Finding
The trigger value at line 25 ("cron") is extremely ambiguous and can collide with system, scheduling, or conversational references to cron jobs. In a skill designed for scheduled automation, this creates a heightened risk of unintended autonomous execution or control actions being invoked merely by discussing scheduling concepts.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.