Security audit

Demo Resource Upload Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward browser automation guide, but users should treat saved browser state, cookies, and storage as sensitive account data.

Install only if you want an agent to drive browser sessions. Do not commit or share saved auth files, cookies, localStorage values, screenshots, or PDFs from logged-in sessions, and prefer test or low-privilege accounts for automation.

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 (2)

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill explicitly recommends saving and loading browser authentication state (`state save auth.json` / `state load auth.json`) but does not warn that these files can contain session cookies and storage tokens that may grant account access if exfiltrated. In an agent-oriented automation context, users may persist auth artifacts to disk or share workspaces, increasing the chance of credential leakage or reuse across sessions.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill documents direct access to cookies and localStorage, including setting values, without any warning about privacy, token exposure, or session integrity risks. In a browser automation skill, these commands can expose secrets or allow session tampering, and an AI agent may use them in unsafe ways if the documentation normalizes unrestricted access.

Static analysis

No suspicious patterns detected.