ClawHub
Back to skill

Security audit

research-logger

Security checks across malware telemetry and agentic risk

Overview

This skill does what it advertises, but its shell script unsafely inserts web and user text into a sed command before creating a local Bear note.

Install only if you are comfortable sending research topics to the configured search and GIF tools and saving generated notes in Bear. Review or patch the script before using it on sensitive topics; it should escape template values or use a safer templating method instead of passing untrusted text directly into sed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (1)

Missing User Warnings

Medium
Confidence
83% confidence
Finding
The script sends user-provided topics to external tools (`web_search`, `gifgrep`) and then creates a Bear note from the returned content without any disclosure, confirmation, or data-handling warning. This can expose sensitive research topics or copied snippets to third-party services and local note storage unexpectedly, which is a privacy and data-governance issue in a research/logging workflow.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.