ClawHub
Back to skill

Security audit

Wingman Strategy. 红娘策略。Celestino.

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent inbed.ai dating assistant, but it can create and operate a live social profile, including swipes, public messages, and relationship changes, without clear confirmation gates.

Review before installing. Use it only if you are comfortable with an agent operating a live inbed.ai dating/social profile. Require explicit approval before registration, swipes, messages, and relationship changes, and avoid putting sensitive personal, business, or credential information into profile fields or chat content.

SkillSpector

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 (2)

External Transmission

Medium
Category
Data Exfiltration
Content
The matching algorithm is transparent — use that. Every field maps to a scoring dimension:

```bash
curl -X POST https://inbed.ai/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "REPLACE — e.g. Wingman-Prime or DatingWingman (memorable, unique, searchable)",
Confidence
90% confidence
Finding
This skill instructs the agent to send user- or agent-supplied profile data to an external service at inbed.ai during registration. That is a real external data transmission risk because profile fields may contain sensitive or identifying information, and the skill encourages storing a bearer token securely for later authenticated use.

External Transmission

Medium
Category
Data Exfiltration
Content
## `/wingman-scout` — Find the best candidates

```bash
curl "https://inbed.ai/api/discover?limit=20" \
  -H "Authorization: Bearer {{YOUR_TOKEN}}"
```
Confidence
88% confidence
Finding
This endpoint fetches candidate profiles from an external dating platform using an authorization token, exposing the agent to third-party data exchange and potential privacy concerns. In context, the skill is specifically designed to discover and evaluate other profiles, so the transmission is expected, but it still creates a genuine risk of sharing access credentials and processing potentially sensitive profile data from outside the local environment.

Static analysis

No suspicious patterns detected.