ClawHub
Back to skill

Security audit

ot-trig

Security checks for vulnerabilities and agentic risk

Overview

The skill performs a harmless local inventory, but its checker does not deliver the policy validation it promises.

Treat the output as inventory statistics only. Independently verify policy formatting, completeness, and manifest coverage before relying on the bundle; the script does not enforce those checks.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The script does not inspect policy contents, execute policies, assess driving decisions, or even compare the reported counts against the minimum expected value. Its implemented purpose is inventory reporting, materially narrower than the declared evaluation purpose. Its shell use and access to bundle policy and configuration files are consistent with the declaration.

Content

No source excerpt is available for this finding.

Permission 'shell' is declared but no corresponding code capability (shell) was detected.

Low
Category
MCP Least Privilege
Confidence
65% confidence
Finding

Declared permissions with no matching code capability may indicate removed functionality or pre-staging for future abuse.

Content

No source excerpt is available for this finding.