ClawHub
Back to skill

Security audit

OPC 提示词工作台

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-template helper with no code execution, persistence, or hidden data access.

Install this if you want a Chinese prompt-template workbench for OPC/business workflows. Treat its emotional-support and contract-review templates as drafting and triage aids only, and redact sensitive contract, client, employee, health, or financial details before sending them to any external model.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill description contains broad trigger phrases such as requests for prompts and prompt optimization, which can match a wide range of ordinary user intents and cause the skill to activate too often. Over-broad activation can route unrelated conversations into this skill, increasing the chance of unintended access to local references or misuse of specialized prompt content outside its intended scope.

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The description is written as Chinese-only and does not indicate whether the skill can adapt to the user's language, which can create unsafe misunderstandings in higher-risk areas like contract review or emotional reflection. If a user interacts in another language, the model may still invoke the skill and produce inaccurate or confusing outputs because locale and language expectations are not clearly bounded.

Natural-Language Policy Violations

Low
Confidence
84% confidence
Finding
The file presents the skill name and description in Chinese while prescribing an English default prompt for invocation. This can create an implicit language policy mismatch without documented user opt-in or an explicit language-selection mechanism.

Natural-Language Policy Violations

Low
Confidence
87% confidence
Finding
This markdown file presents all skill-facing content in a single language and does not provide user opt-in, alternatives, or justification for the locale restriction. Under the policy, forced language or locale constraints can be a natural-language policy violation when no choice is offered.

Static analysis

No suspicious patterns detected.