Security audit

agent-browser

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward browser automation skill, with sensitive cookie and login-state features that users should handle like credentials.

Before installing, understand that this skill helps an agent control a browser and can read or reuse authenticated browser state. Treat saved auth files, cookies, and localStorage output like passwords: keep them out of source control and shared workspaces, minimize what you inspect, redact logs, and delete state files when no longer needed.

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: 88% confidence
Finding: The skill explicitly recommends saving and loading browser auth state but does not warn that these files can contain cookies, tokens, and other session material equivalent to active credentials. In an agent context, this omission increases the chance that sensitive session state will be persisted insecurely, reused across tasks, or exposed through logs, shared workspaces, or source control.

Missing User Warnings

Medium

Confidence: 85% confidence
Finding: The skill documents commands for reading and setting cookies and localStorage without warning that these interfaces may expose session identifiers, CSRF tokens, PII, or other sensitive browser data. Because this skill is designed for AI agents and automation, the lack of guidance makes accidental exfiltration, over-collection, or unsafe handling of browser secrets more likely.

Static analysis

No suspicious patterns detected.