ClawHub
Back to skill

Security audit

cmg gem油藏数值模拟专家

Security checks for vulnerabilities and agentic risk

Overview

This skill is a domain-specific CMG GEM simulation assistant that discloses its local file and simulator-use workflow, with no evidence of hidden, persistent, network, credential, or destructive behavior.

Install only if you expect the agent to create DAT files and run your local CMG GEM simulator. Use a dedicated project directory, choose filenames explicitly, check before overwriting existing work, and confirm simulator runs because they may consume local compute or licensed software resources.

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
  • 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 (2)

Missing User Warnings

Low
Confidence
94% confidence
Finding
The skill explicitly states it will write generated DAT files into the user's working directory, but it does not require explicit per-action confirmation, define safe path constraints, or warn about overwriting existing files. In a local agent setting, this can lead to unintended modification or loss of user data if the working directory or filename is ambiguous, guessed incorrectly, or influenced by user input.

Missing User Warnings

Low
Confidence
96% confidence
Finding
The skill instructs the agent to execute a local simulator executable (`gm202420.exe`) and to iteratively rerun it based on generated inputs, but it provides no explicit safety guardrails about executing local binaries. Running local executables from an agent can trigger unintended code execution, resource abuse, or harmful side effects, especially when execution parameters, working directories, or input files are derived from user-controlled content.

Static analysis

No suspicious patterns detected.