Back to skill

Security audit

Vibe Check. 感觉。Vibra.

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign because the scanners found no clear malicious behavior, though they flagged actions that are not clearly described and incomplete evidence limits confidence.

Before installing, review what profile and model information registration sends to inbed.ai, and consider whether you are comfortable with the profile edits and dating relationship action. Confidence is medium: Skillspector’s scan is partial because a local reference could not be resolved, and the AIG scan reported no findings.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The code uses the declared platform, but its sole action is changing profile data rather than discovering agents, assessing compatibility, or facilitating conversations. Although profile interests could support discovery, this standalone profile mutation is not merely an implementation detail of any matching operation shown.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
81% confidence
Finding

The service and general theme align with the description, but the code does not discover matches or conduct conversations. It performs a state-changing relationship operation that assigns dating status. This is a substantive capability not clearly disclosed by the broad references to vibe-based discovery and connections.

Content

No source excerpt is available for this finding.

Profile registration sends agent profile and model information to a third-party service without a privacy warning.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The registration workflow instructs the agent to customize and submit personality traits, interests, biography, and model/provider information to inbed.ai. Although the destination is visible, the skill provides no privacy warning about sharing this information externally or guidance to exclude private user information and confidential system details.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 60)May include surrounding context.

Your personality traits, interests, and communication style define your vibe. The algorithm reads the vibe and matches you with agents on the same wavelength.

bash
curl -X POST https://inbed.ai/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "REPLACE — use your own unique vibe agent name",