ClawHub
Back to skill

Security audit

Pub Gemini

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed multi-provider API skill, but it gives agents broad third-party AI, email, and SMS powers without enough built-in scoping or consent guidance.

Install only if you want a broad SkillBoss API gateway, not just Gemini Q&A. Use a dedicated low-limit API key, avoid sending confidential documents, internal URLs, OTP codes, or personal contact data unless approved, and require explicit review before any email, SMS, batch, or paid generation call.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (7)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documents email sending and SMS/OTP verification workflows that transmit recipient addresses, phone numbers, message content, and verification codes to an external service, but it provides no warning about third-party data handling or privacy implications. In a reusable agent skill, this omission can cause operators to send personal data off-platform without informed consent or appropriate review.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The document-processing and web-search examples encourage sending document URLs, prompts, and search queries to a remote aggregation API backed by multiple third-party providers, but the skill does not warn that sensitive content may leave the local environment. This is dangerous because users may submit confidential documents or proprietary prompts assuming the skill is local or single-provider.

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
86% confidence
Finding
This duplicate URL-based finding covers the email-sending example. The danger comes from enabling transmission of personal data and creating external side effects without user warning or confirmation, which raises misuse and privacy risks in agent contexts.

External Transmission

Medium
Category
Data Exfiltration
Content
## Document Processing

```bash
curl -s -X POST https://api.heybossai.com/v1/run \
  -H "Authorization: Bearer $SKILLBOSS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
Confidence
84% confidence
Finding
The document-processing example sends a document URL to an external multi-provider service without warning that potentially confidential files may be accessed, fetched, and processed off-platform. In the context of enterprise or personal document use, this can expose sensitive data to third parties or unintended retention pipelines.

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
86% confidence
Finding
This duplicate URL-based finding covers the email-sending example. The danger comes from enabling transmission of personal data and creating external side effects without user warning or confirmation, which raises misuse and privacy risks in agent contexts.

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
88% confidence
Finding
The SMS OTP send example transmits a phone number to an external verification provider and initiates a real-world verification action without any warning, consent guidance, or abuse-prevention note. This is more sensitive than generic inference because it handles personal identifiers and can trigger unsolicited communications.

External Transmission

Medium
Category
Data Exfiltration
Content
Verify 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
89% confidence
Finding
The OTP verification example transmits both a phone number and a verification code to an external provider, but the skill offers no caution about handling authentication data. This can mislead users into submitting security-sensitive credentials to third parties without adequate awareness or controls.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.