ClawHub
Back to skill

Security audit

Working Memory

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local project-memory helper that writes disclosed Markdown files so work can resume across sessions.

Install this if you want the agent to keep a local project memory file. Review WORKING.md before checkpointing sensitive work, and use more explicit prompts if you want to avoid accidental updates from generic words like "checkpoint."

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
92% confidence
Finding
The skill declares very broad trigger phrases such as "checkpoint," "save state," and "resume where we left off," which are common conversational phrases that can be invoked unintentionally in many normal workflows. This can cause the skill to activate outside the user's explicit intent and write or modify WORKING.md/CLAUDE.md unexpectedly, increasing the risk of accidental file changes or persistence of sensitive transient context.

Static analysis

No suspicious patterns detected.