Complaint Handling

• Outcome: Review
• Latest audit: Jun 9, 2026 at 3:02 PM UTC
• Version:
  1.0.0

Overview

This complaint-management skill is coherent, but it handles identifiable customer complaint data with local duplication and optional WeCom forwarding without enough privacy controls or disclosure.

Review before installing. Use this only in an authorized business environment, configure storage paths deliberately, restrict filesystem access to complaint records and generated reports, avoid putting real customer data into broad chat groups, and add redaction, retention, and approval controls before enabling WeCom forwarding.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (5)

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill explicitly references customer identity, complaint records, contact details, and a local spreadsheet data source, but it does not provide clear privacy handling guidance such as data minimization, access control, retention, masking, or user-consent expectations. In a complaint-handling workflow, this increases the risk of unnecessary exposure or misuse of personally identifiable information during storage, reporting, and follow-up.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The code persists complaint records that include personal and sensitive fields such as customer name, room number, complaint content, follow-up status, and group identifiers. Storing this data locally without any visible access controls, minimization, retention policy, encryption, or user notice increases the risk of privacy leakage if the host, files, or backups are accessed by unauthorized parties.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The reporting logic writes JSON reports to local paths containing complaint details, overdue details, customer names, and complaint content. Reports often aggregate and duplicate sensitive data, which broadens exposure and can leak historical customer issues beyond the minimum necessary operational use.

Missing User Warnings

Medium

• Confidence: 80% confidence
• Finding: The backup routine duplicates potentially sensitive complaint records, likely containing personal and service-related data, into a fixed local backup directory without any access-control, encryption, retention, or visibility safeguards. If that directory is accessible to other local users, synced, or included in broader tooling, it increases the exposure surface and can lead to unintended disclosure of duplicated sensitive data.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The module sends complaint records containing customer names, room numbers, complaint content, and timing to an external WeCom webhook, but there is no visible consent flow, data-minimization control, warning, or classification check before transmission. In a complaint-handling context this data may include sensitive personal or business information, so automatic forwarding to a third-party messaging endpoint creates a real confidentiality and privacy risk if the webhook is misconfigured, shared too broadly, or points outside the intended trust boundary.

VirusTotal

65/65 vendors flagged this skill as clean.