ClawHub
Back to skill

Security audit

Chatmosp Input Coordinator

Security checks for vulnerabilities and agentic risk

Overview

This is a bilingual coordinator for ChatMOSP workflows; it has some language-routing/documentation mistakes but no deceptive, destructive, persistent, or credential-seeking behavior.

Install only if you are comfortable with a ChatMOSP coordinator that delegates real simulation and file-generation work to companion skills. Review those companion skills separately, and correct the bilingual routing labels if Chinese/English behavior matters for your use case.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
> - If the user writes in Chinese, read `SKILL.md` in this same directory and respond in Chinese.

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 1)May include surrounding context.

md
---
name: chatmosp-input-coordinator
description: |
  chatMOSP 系统的入口技能。负责解析中英文自然语言输入，识别 MSR/KMC/参数查询

Hidden Instructions

High
Category
Prompt Injection
Confidence
60% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · SKILL_cn.md (reported line 1)May include surrounding context.

md
---
name: chatmosp-input-coordinator
description: |
  chatMOSP 系统的入口技能。负责解析中英文自然语言输入，识别 MSR/KMC/参数查询

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The language-routing logic is inconsistent: for English input it instructs the agent to read SKILL_cn.md, while the current file is itself SKILL_cn.md and later references conflict with the Chinese/English authority mapping. This kind of routing error can cause the model to follow the wrong instruction set or skip relevant safety/behavioral content, creating unreliable behavior in a multi-file skill.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger definition is broad enough that ordinary discussion of catalysis topics could invoke the skill unintentionally. In an agent environment, overbroad activation can cause the system to parse user content as task requests and dispatch downstream skills, increasing the chance of unintended file operations, parameter generation, or simulation workflow execution without clear user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger description is broad enough to overlap with ordinary chemistry-related conversation, which can cause the skill to activate when the user did not intend to launch a computational workflow. In an orchestration skill that routes into parameter building and simulation tools, false activation can lead to unintended task execution, confusing prompts, or unnecessary downstream actions.

Content

No source excerpt is available for this finding.

Ssd 1

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The document tells the model to treat another file as authoritative and to ignore the rest of the current document based on language. Cross-file authority switching increases the risk that important constraints, safety notes, or routing logic in the current file are bypassed, especially when file naming and locale mapping are already inconsistent.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The routing instructions say that Chinese input should read SKILL.md in the same directory, implying this file is the English version. However, the file structure later labels SKILL.md as the Chinese reference and SKILL_cn.md as 'This file (English)', which reverses the expected naming/documentation and creates a direct documentation contradiction about which file corresponds to which language.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.