ClawHub
Back to skill

Security audit

M-Flow Memory

Security checks for vulnerabilities and agentic risk

Overview

This memory skill is coherent, but it can store and index conversation content with unclear consent, retention, deletion, and sharing boundaries.

Review before installing. Use only non-sensitive content unless you have verified where M-Flow stores data, who or what agents can search it, how to delete retained memories, and whether distillation summarizes or stores raw transcripts in your setup.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation criteria are overly broad and can cause the skill to trigger on ordinary memory-related user requests without clear consent or necessity. In a memory skill, unnecessary activation increases the chance that user content is processed, indexed, or retained when the user did not explicitly intend persistent storage or memory operations.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill describes add, memorize, and search operations without warning that conversation content may be stored, indexed, and later retrieved. This creates a meaningful privacy and data-handling risk because users may disclose sensitive information under the assumption it is only being used transiently, while the skill is designed for persistence and distillation.

Static analysis

No suspicious patterns detected.