ClawHub
Back to skill

Security audit

Chatmosp Input Coordinator

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed ChatMOSP workflow coordinator for catalysis simulations, with some broad trigger wording but no evidence of hidden, destructive, or credential-seeking behavior.

Before installing, make sure you actually use the ChatMOSP companion workflow and review the companion skills it dispatches to, since those may create files or run simulations. Use explicit requests such as running MSR/KMC or showing parameters, and confirm generated parameters before allowing downstream execution.

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)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger conditions are broad enough to activate on generic discussion of metals, catalysis systems, or parameter adjustments, which can cause unintended skill invocation. In an agent environment, overbroad activation increases the chance of context hijacking, inappropriate tool routing, or execution of expensive or state-changing workflows when the user only wanted informational discussion.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger text is broad enough to activate on ordinary discussion of catalysis topics, not just explicit requests to run the skill. Over-broad activation can cause unintended routing into downstream skills that create files or launch simulations, violating user intent and increasing prompt-injection exposure from unrelated conversation.

Static analysis

No suspicious patterns detected.