ClawHub
Back to skill

Security audit

Deep Research

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed CellCog deep-research integration, with a caution that users should independently verify high-stakes financial, legal, regulatory, or due-diligence outputs.

Install only if you are comfortable using CellCog as an external research provider with your CELLCOG_API_KEY. For financial, legal, regulatory, medical, investment, or due-diligence work, explicitly request citations and verify key claims against primary sources or qualified professionals before relying on the results.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly promotes investment analysis, financial research, and due diligence use cases but does not warn that outputs may be incomplete, stale, or unsuitable as professional financial advice. Users may over-trust the tool for investing or business decisions, creating foreseeable harm from inaccurate or hallucinated conclusions in a high-stakes domain.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill recommends using enhanced modes for high-stakes due diligence, regulatory compliance, and institutional-grade analysis while emphasizing deeper reasoning and quality gains, but it does not require independent verification. This framing can increase user reliance in costly or compliance-sensitive contexts where model errors, omitted sources, or unsupported conclusions could lead to legal, financial, or operational harm.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.