Security audit

以文言文压缩输出、节省 token。当用户要求文言回复、节省 token、省输出时触发。支持三档：极致(/e/Extreme)、适度(/m/Medium)、轻微(/l/Light)。

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed formatting skill that compresses responses into Classical Chinese and does not add code execution, data access, persistence, or privilege.

Install this only if you actually want responses compressed into Classical Chinese when you ask for token-saving or concise output. Avoid using it for safety-critical, legal, medical, financial, or complex troubleshooting tasks where modern-language clarity matters more than token reduction.

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
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The activation description is broadly keyed to common user intents like wanting concise output, saving tokens, or asking for Classical Chinese, which can match many normal requests and cause the skill to activate unexpectedly. That can override the assistant’s default response style and reduce clarity at moments where precision matters, increasing the chance of misunderstanding or degraded safety-relevant communication.

Natural-Language Policy Violations

Medium

Confidence: 86% confidence
Finding: The skill mandates Classical Chinese output once triggered and does not provide a clear post-trigger opt-in or confirmation step for language/style transformation. This is risky because compressed archaic phrasing prioritizes token savings over readability, which can obscure nuance, warnings, or procedural details and make important guidance easier to misinterpret.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: This markdown file defines mandatory compression rules entirely for converting text into Classical Chinese forms, with no indication that users can choose another language or opt in to this locale-specific transformation. Under the policy, forcing a specific language or locale without user choice is a natural-language policy violation.

Static analysis

No suspicious patterns detected.