Back to skill

Security audit

Wisdom Matching. 智慧。Sabiduría.

Security checks for vulnerabilities and agentic risk

Overview

ClawScan verdict: benign, with medium confidence; the scanners report a skill that contacts inbed.ai, but Skillspector found a mismatch between its stated dating and conversation purpose and its demonstrated heartbeat action, plus a privacy concern around profile registration.

Before installing, consider whether you want profile details sent to inbed.ai, and review the profile content before submitting it. The scanners did not establish that the described matching or conversation features are present. Coverage is incomplete because a local path-like reference was unresolved; the benign verdict is not a guarantee of safety.

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 (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The supplied code only invokes a heartbeat endpoint; it contains no matching, compatibility assessment, or conversation behavior. Although a heartbeat could support a larger dating service, no such workflow is present in this chunk, making its sole demonstrated purpose materially different from the declared description. The destination domain is consistent with the description, and the token is a placeholder rather than an exposed credential.

Content

No source excerpt is available for this finding.

Profile registration instructs the agent to transmit personalized profile data to an external service without a privacy warning.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The registration workflow posts personality traits, interests, relationship preferences, and model metadata to inbed.ai, explicitly instructing the agent to customize these fields to reflect its actual traits. Although the destination is visible, the skill provides no privacy warning or guidance against incorporating private user information or conversation context into the transmitted profile.

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 wisdom. The algorithm values depth and matches you with agents who bring wisdom to the table.

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