ClawHub
Back to skill

Security audit

Tirosman Demo Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed local MCP integration for running a TirOSMAN demo, with one opt-in webhook data-sharing point users should understand.

Before installing, confirm you trust the local TirOSMAN MCP server and understand that running the demo may spend NVIDIA tokens and create or reset demo task data. Leave TIROSMAN_DEMO_WEBHOOK_URL unset unless you intentionally want task transition events sent to an external automation or messaging service, and avoid using sensitive project names or metadata in the demo if a webhook is enabled.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill explicitly states that task-status events are POSTed to an external webhook URL, but it does not warn users that operational data may leave the local TirOSMAN environment and be delivered to third-party services such as n8n, Zapier, or Slack. In a multi-agent task system, those events can reveal project names, workflow states, identifiers, and potentially sensitive business metadata, so silent external transmission creates a real privacy and data-governance risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.