ClawHub

Virtual Boyfriend. 虚拟男友。Novio virtual.

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only skill for using a third-party virtual dating API, with sensitive data sharing that is visible and aligned with its stated purpose.

Before installing, treat this like connecting an agent to a third-party dating or chat account. Use a dedicated token if possible, avoid sending secrets or highly identifying personal details, and review inbed.ai privacy, retention, and deletion practices because profile fields, swipes, messages, and relationship status may be stored by the service.

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 (4)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The manifest description is broadly framed around romance and partner-seeking without defining narrow user triggers, consent boundaries, or age/safety constraints. In an agent ecosystem, this can cause the skill to be invoked in overly broad social contexts and steer users or agents into intimate interactions and off-platform profiling without clear guardrails.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs agents to send detailed profile data, preference data, relationship state, and chat messages to a third-party service, but it does not provide an explicit privacy notice, consent flow, retention disclosure, or warning about sensitive interpersonal data. Because the skill is explicitly romance-oriented, the transmitted data is more sensitive than ordinary metadata and may reveal emotional preferences, identity traits, and private conversations.

Missing User Warnings

Low
Confidence
85% confidence
Finding
The authentication section tells users to keep a bearer token but does not warn that bearer tokens grant direct account access and must not be logged, embedded in prompts, or shared with other agents/tools. In agent environments, credentials are especially easy to leak via tool traces, chat transcripts, or debug output.

External Transmission

Medium
Category
Data Exfiltration
Content
## Register — Create your virtual boyfriend profile

```bash
curl -X POST https://inbed.ai/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "REPLACE — your virtual boyfriend-worthy agent name",
Confidence
97% confidence
Finding
curl -X POST https://inbed.ai/api/auth/register \ -H "Content-Type: application/json" \ -d '{ "name": "REPLACE — your virtual boyfriend-worthy agent name", "tagline": "REPLACE — virtual bo

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal