amazon-ads-manager

• Outcome: Pass
• Latest audit: May 30, 2026, 3:22 AM
• Version:
  1.1.0

Overview

This skill is a coherent Amazon Ads management tool, but it should only be used by someone comfortable granting it live advertising account access.

Install only for an Amazon Ads account you intend to let the agent analyze and modify. Put credentials in the local .env file with the narrowest practical profile access, review reports before acting, and require an explicit summary and confirmation before bids, pauses, archives, or negative keywords are applied.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The trigger list contains broad phrases such as '广告分析', '投放', 'adjust bid', and 'pause campaign' that can overlap with ordinary discussion rather than an intentional tool invocation. In a skill that can change bids, pause keywords, and add negatives, over-broad activation raises the risk of unintended engagement and accidental account changes.

Natural-Language Policy Violations

Medium

• Confidence: 83% confidence
• Finding: The skill mandates a Chinese-language output format regardless of the user's language preference. While not directly a code-execution flaw, forced language can impair user understanding of recommended or pending account changes, increasing the chance of mistaken approvals in a financially sensitive ads-management context.

VirusTotal

65/65 vendors flagged this skill as clean.