Security audit

backup-optimizer

Security checks for vulnerabilities and agentic risk

Overview

This skill only generates a plan, but the plan can direct users to bundle credential files into deployment artifacts, which creates a real secret-leakage risk.

Review before installing. Do not use this skill for real deployment planning unless its rules are changed to exclude secrets by default and refer to secret managers or runtime injection instead. Treat any generated plan as sensitive because it may list where credentials live.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Description-Behavior Mismatch

Medium

Confidence: 98% confidence
Finding: The skill explicitly groups environment files, credentials, and configuration into deployment bundles and even states that service account keys are needed at runtime. Packaging secrets as deployment artifacts increases the chance of credential leakage, broader artifact distribution, and long-lived secret exposure across environments and storage systems.

Context-Inappropriate Capability

High

Confidence: 99% confidence
Finding: The code's purpose is framed as packaging optimization, but it deliberately plans inclusion of credential files and runtime secrets, which is a sensitive and risky capability beyond ordinary packaging needs. In the context of a deployment helper, this is especially dangerous because users may trust the generated plan and propagate secrets into build artifacts, backups, or shared repositories.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The generated packaging_plan.json records paths to sensitive environment and credential files without warning or protective handling. Even if contents are not written, the plan becomes a roadmap to secret locations and may be stored, logged, or shared as part of build outputs, increasing exposure and easing follow-on compromise.

Ssd 3

High

Confidence: 98% confidence
Finding: The skill explicitly instructs downstream packaging to include credential material such as .pem, .key, and service-account.json files in deployment bundles. Even though this skill only generates a plan, normalizing secrets as deployable artifacts significantly increases the chance of credential leakage through build artifacts, logs, caches, artifact repositories, and broader access by deployment systems.

Static analysis

No suspicious patterns detected.