ClawHub
Back to skill

Security audit

session-management

Security checks for vulnerabilities and agentic risk

Overview

This skill is a transparent session-log and project-state workflow; it writes local project memory files as its core function but does not show hidden, destructive, or exfiltrating behavior.

Install this only in projects where you want Claude to maintain `state/` and `sessions/` Markdown records. Review generated summaries before `/end` or `/update` writes, and approve Claude memory additions only when you want that information reused across future project conversations.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill repeatedly instructs the agent to append to and modify multiple project files (`state/*.md`, `sessions/*.md`, and potentially memory files) as part of normal command execution, but the top-level description does not clearly warn the user up front that invoking the skill causes writes. In an agent setting, hidden or easy-to-miss file modification behavior can lead to unintended persistence, accidental overwrites, and user surprise, especially because the skill also includes auto-created files and always-update steps.

Static analysis

No suspicious patterns detected.