ClawHub
Back to skill

Security audit

Dingtalk Docs Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed DingTalk Docs integration that can change cloud documents, but its artifacts require user confirmation for write actions and do not show hidden or unrelated behavior.

Install only if you intend to let your agent access and manage DingTalk Docs with your account permissions. Treat the MCP URL like a credential, confirm document titles and locations carefully before overwrites or deletes, and avoid installing it in projects where local config files might be accidentally committed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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 (1)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The eval endorses overwriting a remote document from a local file without any user-facing warning, confirmation, or backup step. Because overwrite is destructive, an agent following this pattern could irreversibly replace important cloud content due to ambiguity, stale search results, or mistaken document selection.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.