ClawHub
Back to skill

Security audit

Agentmemory Rest Api

Security checks across malware telemetry and agentic risk

Overview

This is a small documentation-only skill for using a local memory REST API, and its sensitive behavior is disclosed and aligned with that purpose.

Before installing, use this only with a memory server you control. Set AGENTMEMORY_SECRET if other local users or processes may be untrusted, and avoid saving secrets or sensitive workflow context unless the memory server is secured appropriately.

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
96% confidence
Finding
The skill explicitly documents unauthenticated write access to a local memory service and provides a ready-to-run example that persists data without warning about trust boundaries, persistence, or the risks of local cross-process abuse. 'Localhost' is not a meaningful security boundary on a shared host or against other local software, so this can enable unauthorized memory poisoning, persistence abuse, or leakage of sensitive workflow context if operators assume the interface is inherently safe.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.