ClawHub

Amazon Analyse

Security checks across malware telemetry and agentic risk

Overview

This looks like a legitimate Amazon competitor-analysis skill, but it includes overbroad third-party API references with account-changing keyword actions that are not clearly needed for reporting.

Install only if you trust Sorftime with your product research queries and API key. Treat the generated reports as persistent business records under reports/. Restrict use to the read-only Amazon analysis endpoints unless you explicitly want the agent to modify keyword-library data or use the broader TikTok/1688 research APIs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The referenced MCP documentation includes write-capable operations for a user's Amazon keyword favorites/library such as add, move, and delete. These capabilities are unrelated to the stated skill purpose of competitor listing analysis, so exposing or retaining them increases the risk of unintended account changes, privilege overreach, or prompt-induced misuse if the agent can call broader tools than necessary.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill states that it will automatically save analysis results into `reports/` without clearly surfacing this local write side effect to the user at invocation time. Silent filesystem writes can surprise users, persist potentially sensitive business data, and create audit or data-retention issues in environments where agents should not write files without explicit consent.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The document provides an authenticated example request to a third-party endpoint using an API key in the URL, but it does not disclose that user-supplied product, keyword, category, or market data will be transmitted off-platform. In the context of an agent skill that automates competitor analysis and report generation, this omission creates a meaningful data-exposure risk because operators may unknowingly send sensitive business queries or proprietary research inputs to an external service.

VirusTotal

34/34 vendors flagged this skill as clean.

View on VirusTotal