Alibabacloud Dataworks Data Governance

• Outcome: Review
• Latest audit: May 29, 2026 at 9:34 PM UTC
• Version:
  0.0.1-beta.1

Overview

This skill is mostly a legitimate Alibaba Cloud DataWorks tag-management helper, but it needs review because its setup guidance and triggers are broader than the narrow cloud metadata task.

Install only if you intend to let an agent work with Alibaba Cloud DataWorks data-governance tags. Use a dedicated least-privilege RAM user or role, avoid admin or unrelated service plugins, do not paste access keys into chat or commands, and manually confirm every create, update, bind, or unbind action with the exact region, project, assets, and tags.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (5)

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The file is a full-purpose Aliyun CLI installation and administration guide, not a narrowly scoped reference for DataWorks data-governance tag management. In an agent skill, this broadens operator capability far beyond the declared purpose and can facilitate unrelated cloud actions if the agent or user follows the guide verbatim.

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: This section teaches multi-profile and cross-service CLI usage patterns that enable broader cloud administration unrelated to tag management. In the context of an agent skill, such capability expansion increases the chance of misuse, privilege creep, or an agent being steered into operating on unrelated Alibaba Cloud resources.

Context-Inappropriate Capability

Low

• Confidence: 84% confidence
• Finding: The next-steps section encourages installing unrelated plugins and exploring other cloud services, which extends the skill's operational surface beyond its declared DataWorks tagging purpose. While likely written as general helpful documentation, it can normalize broader cloud access and distract from least-functionality principles.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The trigger list includes broad phrases such as 'tag management', 'query data assets', and 'list data assets', which are generic enough to match ordinary user requests outside the intended Alibaba Cloud DataWorks context. This can cause unintended skill invocation and execution of sensitive tag-management or asset-query workflows against real cloud resources, increasing the risk of accidental metadata changes or disclosure of asset inventory.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill supports unbinding tags from data assets, which is a state-changing metadata operation, but the documentation does not present a clear user-facing warning that this modifies governance metadata and may affect discovery, classification, policy workflows, or downstream automation. In a data governance context, silent or under-signaled metadata changes are risky because operators may treat the action as harmless cleanup when it can alter compliance or operational controls.

VirusTotal

65/65 vendors flagged this skill as clean.