AWS S3 Bucket Audit

• Outcome: Pass
• Latest audit: May 28, 2026, 11:00 AM
• Version:
  1.0.0

Overview

This is a coherent AWS S3 security-audit skill, but users should treat its remediation snippets as live cloud-configuration changes.

Install only if you intend to audit AWS S3 security posture with authorized AWS credentials. Run the read-only checks first, verify each target bucket's intended public access model, and treat any put-* AWS CLI or boto3 examples as live changes that should go through normal change control, especially for static websites, CloudFront origins, partner integrations, encryption/KMS settings, and production buckets.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The skill is presented as an auditing guide but includes ready-to-run remediation commands that change bucket security posture. This is dangerous because an operator or autonomous agent expecting read-only assessment may execute write operations that alter production access, encryption, or versioning, potentially causing outages or unreviewed configuration drift.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The skill is described as an auditing reference, but the included example performs a state-changing operation by calling put_public_access_block(). In an audit context, readers may expect examples to be read-only; including a mutating command without strong labeling can cause unintended configuration changes in production buckets and blurs the boundary between assessment and remediation.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The remediation commands perform direct changes to S3 bucket public access, encryption, and versioning without any confirmation, dry-run equivalent, or precondition checks. In a cloud-production context this can cause service disruption, break intended public websites or partner integrations, and apply incorrect KMS settings if executed mechanically by users or agents.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The example changes S3 Public Access Block settings without an explicit warning that it modifies bucket security configuration. Even though the specific change is security-hardening, undocumented write operations in reference material can mislead users into running commands they believe are informational, creating operational risk and unexpected side effects.

VirusTotal

66/66 vendors flagged this skill as clean.