Yunlv Email Writer

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:44 PM UTC
• Version:
  1.0.5

Overview

This skill is a disclosed local business-email drafting helper, with no evidence of hidden sending, credential access, exfiltration, or destructive behavior.

Before installing, verify the publisher identity because the manifest has inconsistent author/homepage metadata. Treat generated emails as drafts, review pricing, payment terms, legal/compliance issues, and anti-spam obligations before sending, and avoid storing sensitive customer data in the disclosed local draft/log folders unless that is acceptable for your environment.

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 (4)

Vague Triggers

Low

• Confidence: 83% confidence
• Finding: Broad trigger phrases like 'email template', 'B2B email', and 'quotation email' can cause the skill to activate in contexts where the user did not intend to invoke this specific capability. Unintended invocation can expose user-provided business content to the skill's drafting workflow and increase the chance of inappropriate data retention or cross-context handling.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill explicitly defines persistent local storage paths for drafts, sent records, templates, sequences, and logs, but does not clearly require informed user consent, retention limits, or deletion controls. Business emails often contain customer identities, pricing, terms, and contact data, so silent or unclear retention increases privacy and confidentiality risk if the host system is shared, compromised, or backed up insecurely.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger "邮件撰写" is broad enough to match many ordinary email-writing requests outside the stated foreign-trade/B2B scope. This can cause unintended skill activation, routing users into a narrower or mismatched workflow and increasing the chance of inappropriate prompt injection or policy bypass through over-selection of the skill.

Vague Triggers

Low

• Confidence: 80% confidence
• Finding: The trigger set and description describe foreign-trade/business email assistance, but they do not define boundaries that separate this skill from general email drafting. Without exclusions or disambiguation rules, the orchestrator may invoke this skill for broader requests than intended, reducing reliability and potentially exposing users to incorrect domain-specific outputs.

VirusTotal

61/61 vendors flagged this skill as clean.