ClawHub
Back to skill

Security audit

context-resume

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed context-resume helper that reads project memory files at session start, with the main caution that those files may contain private work notes.

Install this if you want automatic continuity across sessions and are comfortable with the agent reading workspace memory files, daily notes, task status files, and project docs when a session begins. Keep sensitive secrets out of those memory documents, or invoke the skill only in workspaces where that context is appropriate.

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
72% confidence
Finding
The 'Session startup (always)' trigger encourages automatic execution at the beginning of every session, which can cause the agent to read memory logs, active task files, and project documents without a task-specific need. In a skill designed to aggregate durable memory, overbroad auto-invocation increases the chance of unnecessary exposure of sensitive workspace data and can expand the scope of information accessed beyond the user's immediate request.

VirusTotal

40/40 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.