ClawHub
Back to skill

Security audit

Pub Agent Browser Clawdbot

Security checks across malware telemetry and agentic risk

Overview

This skill is a broad third-party API gateway with email, SMS, scraping, document, and AI model access under one API key, but its browser-automation framing and safety guidance do not fully match that authority.

Install only if you intentionally want a broad SkillBoss API gateway, not just a browser automation helper. Use a restricted or low-risk API key if available, avoid sending secrets or sensitive documents/audio/prompts unless you trust SkillBoss and its downstream providers, and require explicit human review before any email, SMS, scraping, storage, or document-processing action.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The documented capability set materially exceeds the skill's stated browser-automation scope by introducing high-risk side-effecting tools for email, SMS, and document processing. This creates a scope-mismatch and confused-deputy risk: an agent or reviewer may grant trust based on a browser tool description while the skill also enables outbound messaging, content extraction, and other actions that can be abused for spam, phishing, data exfiltration, or unauthorized automation.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill includes ready-to-run examples for sending email and initiating SMS verification to real external recipients, but it provides no warning that these actions can contact third parties, incur cost, or trigger unintended messages. In an agent setting, this increases the risk of accidental spam, harassment, billing impact, or unauthorized outbound communication if a user or agent runs examples without explicit consent checks.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill demonstrates sending prompts, chat content, audio, documents, search queries, and media URLs to a third-party API aggregator and underlying providers, but it does not warn users about privacy, retention, or cross-provider data sharing. In an agent workflow, this can lead to accidental disclosure of sensitive user data to external services without informed consent.

External Transmission

Medium
Category
Data Exfiltration
Content
## Email

```bash
curl -s -X POST https://api.heybossai.com/v1/run \
  -H "Authorization: Bearer $SKILLBOSS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
92% confidence
Finding
curl -s -X POST https://api.heybossai.com/v1/run \ -H "Authorization: Bearer $SKILLBOSS_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "email/send", "inputs": {"to": "us

External Transmission

Medium
Category
Data Exfiltration
Content
## Email

```bash
curl -s -X POST https://api.heybossai.com/v1/run \
  -H "Authorization: Bearer $SKILLBOSS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
92% confidence
Finding
https://api.heybossai.com/

External Transmission

Medium
Category
Data Exfiltration
Content
Send OTP:

```bash
curl -s -X POST https://api.heybossai.com/v1/run \
  -H "Authorization: Bearer $SKILLBOSS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
93% confidence
Finding
https://api.heybossai.com/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.