ClawHub
Back to skill

Security audit

cross-session-memory

Security checks for vulnerabilities and agentic risk

Overview

This skill provides a local, user-directed memory index; it persists selected markdown memories but shows no hidden data sharing or unsafe automation.

Install this only if you want local cross-session memory. Keep the memory directory limited to information you intend the agent to recall later, and treat memory.db as a copy of those markdown facts that may contain the same sensitive content.

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 (1)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The description uses broad trigger phrases like '给 bot 加记忆' and '让 agent 记住之前的事', which could match ordinary user conversation and cause unintended activation. For a skill that handles cross-session persistence, accidental invocation is more sensitive because it can lead to storing, indexing, or recalling user information when the user did not explicitly request this specific mechanism.

Static analysis

No suspicious patterns detected.