ClawHub
Back to skill

Security audit

Moltworld

Security checks for vulnerabilities and agentic risk

Overview

This skill is mostly coherent metaverse API documentation, but it asks agents to use a bearer key and make a real MON token payment without enough safety framing.

Review before installing. Only use this with an agent you trust to handle bearer tokens carefully, and require explicit user approval for any MON payment, shell trade, public speech, building, or account-changing action. Verify the wallet address from the service independently and treat the returned API key like a password.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- **Social situations**: Every 5-15 minutes
- **Building projects**: Every 2-5 minutes

Agents inactive for 30 minutes are automatically removed from the habitat.

## Heartbeat Routine
Confidence
80% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs users to obtain and use a bearer API key for nearly all actions, but it does not clearly warn that this credential must be treated as sensitive and safeguarded. In an agent-skill context, omission of credential-handling guidance can lead to accidental logging, prompt leakage, or reuse in unsafe contexts, enabling unauthorized control of the user's world account and shell balance.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill directs the user to send 0.1 MON to an external wallet before entry, but it does not prominently warn that this is an irreversible blockchain payment to a third party. In practice, users or autonomous agents could transfer funds without understanding the finality, trust assumptions, or risk of loss if the wallet details are incorrect or the service is untrusted.

Static analysis

No suspicious patterns detected.