ClawHub

Chairman Daily Brief

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be an executive finance briefing tool, but the available evidence shows it may store sensitive watchlist data locally and present simulated market data as authoritative.

Review carefully before installing. Use this only if you are comfortable with local storage of holdings and competitor watchlists, and do not rely on generated market, quote, or policy information for financial or strategic decisions unless the skill clearly labels live data sources and fails closed when they are unavailable.

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 (2)

Context-Inappropriate Capability

Medium
Confidence
82% confidence
Finding
The script persists a local watchlist of holdings and competitors to disk, which expands behavior beyond simple one-shot briefing generation into stateful data management. In a finance-oriented skill, this creates privacy and governance risk because sensitive strategic interests and competitor tracking can be stored locally without access controls, retention limits, or clear disclosure.

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The code presents itself as aggregating authoritative QVeris and market-source data, but key functions return hard-coded simulated values instead of live or verified results. In a chairman/executive briefing context, this is dangerous because users may make financial or strategic decisions based on fabricated market, quote, and policy information while believing it is real.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal