Back to skill

Security audit

Drunk Mode

Security checks for vulnerabilities and agentic risk

Overview

This skill only toggles a disclosed playful response style and stores a small local state file, with no evidence of hidden code, network use, credential access, or destructive behavior.

Install only if you want a persistent Chinese-triggered persona toggle. Be aware that casual phrases may enable it accidentally, and use '关闭醉酒模式' to turn it off if responses start using the tipsy style.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger set includes broad conversational phrases like “醉了” and “有点晕”, which can appear in ordinary user dialogue and unintentionally activate the skill. Because activation changes persistent behavior via a state file, accidental triggering can alter subsequent responses without the user clearly intending to enable the persona.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The activation and control instructions are defined only with Chinese trigger phrases, such as "开启醉酒模式" and "关闭醉酒模式", without indicating user opt-in for language or that the skill is intentionally limited to a Chinese-language environment. This can violate language/locale policy when a skill imposes a specific language without choice or justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The file documents that state is stored in ~/.openclaw/workspace/skills/drunk-mode/drunk_state.json and later instructs writing that file, but it does not clearly warn users that enabling the skill persists configuration on disk. Because this affects local user data/state, the skill description should disclose that behavior more explicitly.

Content

No source excerpt is available for this finding.