ClawHub
Back to skill

Security audit

Tirosman Demo Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed controller for a local TirOSMAN demo, with expected token use, demo data changes, bearer auth, and optional webhook forwarding.

Install this only when you trust the local TirOSMAN MCP server and understand the demo can spend model tokens and change demo task records. Run the estimate first, use auto_run only when you intend execution to start, verify the company_id before reset or QA actions, and set TIROSMAN_DEMO_WEBHOOK_URL only to trusted destinations because task transition data will be forwarded there.

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
Confidence
94% confidence
Finding
The skill explicitly states that task-status events are POSTed to an externally configured webhook URL, but it does not warn the user that operational data will leave the local TirOSMAN environment. In an agent skill, this can cause unintentional disclosure of project names, task metadata, workflow state, or other sensitive business context to third-party services such as Slack, Zapier, or n8n.

Static analysis

No suspicious patterns detected.