ClawHub
Back to skill

Security audit

context-manager

Security checks across malware telemetry and agentic risk

Overview

This is a local SQLite memory-management skill with expected persistence and no evidence of network access, exfiltration, or implemented destructive behavior.

Install only if you want a local memory store. Avoid adding secrets, credentials, or regulated personal data, and verify the database location printed by --init. Treat the unimplemented documented commands as unreliable until the publisher updates the script or documentation.

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 (1)

Missing User Warnings

Medium
Confidence
87% confidence
Finding
Documenting destructive operations like delete and reset without explicit warnings increases the risk of accidental irreversible data loss, especially in a memory-management skill handling persistent SQLite data. In agent environments, users may invoke commands verbatim from documentation, so omission of safety guidance can directly lead to unintended destruction of stored context.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.