ISO20000认证助手/ISO20000-certificate-assistant

• Outcome: Pass
• Latest audit: Jun 11, 2026 at 4:55 PM UTC
• Version:
  1.0.1

Overview

This skill locally analyzes and generates ISO 20000 compliance documents, with expected document-parsing risks but no evidence of hidden exfiltration, credential access, destructive behavior, or persistence beyond local outputs and logs.

Install only if you are comfortable letting the skill parse client ISO/ITSM documents locally and include extracted content in analysis or generated drafts. Prefer .docx or .txt for sensitive work, avoid uploading unnecessary secrets or personal data, and treat legacy .doc conversion as higher risk unless the local converter is patched and sandboxed.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Context-Inappropriate Capability

Medium

• Confidence: 82% confidence
• Finding: The skill processes untrusted `.doc` files by launching an external utility (`antiword`). Even without shell injection, this expands the attack surface to vulnerabilities in the installed system binary and can expose the host to parser bugs, crashes, or malicious document-triggered exploitation; in an agent skill that may handle user-supplied documents, that context makes the risk more meaningful.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill processes uploaded client documents, extracts their contents, and uses that content to auto-fill generated documents, but it does not clearly warn users that sensitive business information may be ingested and reproduced in outputs. In a certification-assistance context, uploaded files are likely to contain confidential operational procedures, supplier details, and personal or internal data, so insufficient disclosure increases the risk of unintended data exposure and privacy violations.

Missing User Warnings

Low

• Confidence: 81% confidence
• Finding: The skill states that .doc files may be auto-converted via LibreOffice or antiword, but it does not disclose that third-party local utilities will process the uploaded files. This matters because document conversion tools expand the attack surface for malformed file exploits and may handle files in temporary locations or with different parsing behavior than users expect.

VirusTotal

65/65 vendors flagged this skill as clean.