ClawHub
Back to skill

Security audit

Threads Keyword Search

Security checks across malware telemetry and agentic risk

Overview

This skill is a focused Threads search helper that reads public search-page data and returns matching posts, with only a limited operational-notes file as a minor retention consideration.

Installers should be comfortable with a browser automation skill that reads public Threads search results and runs a local Python helper to generate page-reading JavaScript. Review or periodically clear the optional memory file if you do not want local operational notes retained across runs.

SkillSpector

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

Context-Inappropriate Capability

Low
Confidence
92% confidence
Finding
The skill instructs the agent to read from and append to a persistent memory file containing prior execution experiences. Even though it says not to store keywords or results, this still creates a cross-task state channel unrelated to the immediate user response and can leak operational context, prior user activity patterns, or environment details into future runs. In a security-sensitive agent setting, persistent side-channel memory should be treated as data retention risk unless tightly scoped and sanitized.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.