Security audit

Setv.Wang

Security checks for vulnerabilities and agentic risk

Overview

This skill matches its advertised ad-account purpose, but it should be reviewed carefully because it saves reusable ad-platform login cookies locally in plaintext.

Install only on a trusted machine and only if you accept that ad-platform session cookies will be saved locally in plaintext. Treat the stored data like passwords: restrict filesystem access, avoid shared or backed-up workspaces, delete saved accounts when no longer needed, and prefer a version that encrypts cookies or uses the OS credential store with explicit consent before saving sessions.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The README explicitly promotes saving authentication cookies locally for future automatic login, but provides no warning about the sensitivity of those cookies or the risks of local compromise, theft, or reuse. In the context of ad-platform account management, stolen cookies could allow unauthorized access to advertising accounts, budgets, billing-related data, and campaign controls without needing credentials or MFA.

Vague Triggers

Medium

Confidence: 74% confidence
Finding: The invocation description is broad enough to match generic account-management or login requests, which can cause the skill to activate in situations beyond ad-platform administration. In a skill that performs browser automation and manages login state, overbroad triggering increases the chance of credential-handling actions being invoked unexpectedly or in the wrong context.

Missing User Warnings

High

Confidence: 96% confidence
Finding: The skill explicitly states that login cookies are saved locally but provides no warning, consent flow, retention policy, or protection details for those credentials. Cookies often function as bearer tokens for authenticated sessions, so local persistence can enable account takeover or unauthorized ad-account access if the host is compromised, shared, or improperly permissioned.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The code serializes and stores full advertising account cookie data in plaintext JSON on disk. Session cookies are authentication secrets; if the local filesystem, backups, logs, or shared workspace are accessed by another user or process, an attacker could reuse them to hijack ad platform sessions and manage billing-sensitive accounts.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: This helper persists authenticated browser cookies to disk in plaintext JSON without encryption, access controls, expiry handling, or any user-facing warning that active session tokens are being stored. In the context of an ad-account-management skill that logs into advertising platforms via cookie reuse, theft of this file could let another party hijack authenticated sessions and access or manipulate ad accounts.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The code collects authentication cookies after login and stores them for later one-click login, but the user-facing flow does not clearly disclose that persistent authentication data will be saved. Session cookies are highly sensitive credentials; if stored insecurely, accessed by other components, or retained longer than expected, they can enable account takeover of advertising platforms.

Static analysis

No suspicious patterns detected.