ClawHub
Back to skill

Security audit

Wanpaike Opc Investor Due Diligence

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a coherent investor due-diligence guide, but one reference file contains an undisclosed write-file tool-call block that could cause unintended repository changes.

Review before installing. The due-diligence templates and public-source research guidance are broadly aligned with the skill description, but the publisher should remove the embedded write_file/tool_call block from the reference file so the skill cannot accidentally instruct an agent to modify local files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
98% confidence
Finding
The reference file is supposed to be passive due-diligence guidance, but it embeds an explicit file-writing tool call that instructs creation of new repository content. In an agent setting, executable-looking instructions inside trusted reference material can cause unintended repository modification, prompt injection, or workflow deviation if the agent treats the reference as operational guidance rather than inert documentation.

Description-Behavior Mismatch

Medium
Confidence
98% confidence
Finding
This content crosses from documenting investor due diligence into instructing the agent to modify the repository by creating a new reference file. That boundary violation is dangerous because skills often ingest local markdown as trusted context, so an attacker or careless author can smuggle state-changing actions into documents that appear informational.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The markdown includes a file-writing tool call without any explicit user-facing confirmation or warning that repository files may be changed. In the context of an agent skill, this increases the risk of silent side effects, because the agent may execute modifications based on embedded content that the user did not separately authorize.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.