Hwp Batch Convert Repo

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a legitimate HWP batch-conversion helper, but it includes an option to automatically approve Hancom security prompts during automation.

Install only if you understand and need unattended Hancom HWP conversion. Keep the auto-allow dialog option disabled for untrusted documents, and only enable it in a controlled run where bypassing Hancom security prompts is acceptable.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The skill contains explicit logic to detect Hancom security dialogs and automatically click '허용/모두 허용' when text matches an access-attempt prompt. In context, this bypasses a security control intended to require user consent before document or automation access, which is broader and more dangerous than routine batch conversion.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The README explicitly documents an option to automatically click through a security confirmation dialog in Hancom HWP automation. Even with the claimed whitelist constraints, bypassing a security prompt removes an intentional trust boundary and can allow document access or automation actions to proceed without meaningful user consent, especially in a batch-processing workflow where many files may be handled unattended.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The document explicitly describes logic to automatically click a security confirmation dialog during HWP COM automation, which undermines a security boundary intended to require human approval. Even with a narrow whitelist, this creates a dangerous precedent of auto-approving access prompts and can enable unintended document access or unsafe automation flows if dialog text, context, or product behavior changes.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The `--auto-allow-dialogs` option can silently approve security prompts during execution without a strong runtime warning or explicit per-event user acknowledgment. That can cause users to unknowingly bypass Hancom's security consent mechanism while processing untrusted documents in bulk.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal