ClawHub
Back to skill

Security audit

GROMACS Skills

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese-language GROMACS command reference with examples and cautions, and it does not install code, persist, or access unrelated data.

Install this only if you want GROMACS help, likely in Chinese. Review any generated commands or run scripts before executing them, especially mdrun production jobs, because simulations can consume significant compute and write or overwrite output files.

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 (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill description is very broad and says the skill should be invoked whenever an agent needs to execute GROMACS commands but is unsure of usage, without defining tighter boundaries such as read-only guidance versus command generation versus execution planning. In practice, broad routing can cause over-invocation and increased exposure to risky operational guidance, especially because the skill includes runnable shell examples and production-run script generation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
78% confidence
Finding

The manifest description is written entirely in Chinese and does not indicate that language selection depends on user preference or that the skill is region-specific. Under the policy, forcing a specific language without opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

该文档的标题和正文完全以中文呈现，但没有说明这是面向特定中文用户群的区域性文档，也没有提供语言选择或英文等替代说明。根据语言/区域政策，若技能内容强制特定语言而无用户选择或明确理由，可构成自然语言策略问题。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest says the skill covers a specific set of GROMACS commands and categories, emphasizing commands like pdb2gmx, grompp, mdrun, trjconv, energy, rms, covar, make_ndx, select, genion, xpm2ps, check, and wham. This file additionally documents commands not declared in the manifest, including x2top, tune_pme, trjorder, dump, principal, do_dssp, order, rotacf, and dielectric, which broadens the actual documented capability beyond the stated description.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown file presents its key warning and the remainder of the content in Chinese, with no indication that another language is available or that the locale choice is intentional. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.