ClawHub

Memory Memoria

Security checks across malware telemetry and agentic risk

Overview

This is a coherent long-term memory skill, but users should understand it can persist and later reuse personal or project context.

Install only if you want OpenClaw to keep durable cross-session memory. Verify the external Memoria plugin source, use dedicated database/API credentials, avoid storing secrets or highly sensitive personal data, and periodically review or purge stored memories.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list contains broad natural-language phrases such as "remember this" and "continue from last time" that are likely to appear in normal conversation, increasing the chance of accidental skill activation. In a memory-management skill, unintended invocation can cause retrieval, storage, correction, or deletion actions against durable memory when the user did not explicitly mean to invoke the plugin.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly encourages storing durable user preferences, profile facts, project decisions, and other long-lived context, but provides no privacy notice, consent boundaries beyond broad utility, retention limits, or data-minimization safeguards. In a memory skill whose purpose is cross-session persistence, this omission increases the chance of collecting and retaining personal or sensitive information longer than necessary, creating privacy and compliance risk.

Natural-Language Policy Violations

Medium
Confidence
97% confidence
Finding
The instruction 'Do not store secrets unless the user explicitly wants them remembered' permits retention of secrets in durable memory based solely on a user request, without any compensating controls such as warnings, redaction, secure handling requirements, or safer alternatives. Because this skill is specifically designed for long-term cross-session memory, storing API keys, passwords, tokens, or other secrets could lead to later disclosure, misuse, or unintended propagation across tasks.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The setup instructions include a credential-bearing environment variable (`MEMORIA_EMBEDDING_API_KEY='sk-...'`) inline with shell commands and provide no warning about secret handling. This can normalize unsafe practices such as pasting real API keys into shell history, shared terminals, logs, screenshots, or documentation, increasing the chance of credential disclosure.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal