ClawHub
Back to skill

Security audit

Personal Assistant

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple local daily-briefing generator with some disclosure gaps around file output and optional automation, but no evidence of hidden, deceptive, exfiltrating, or destructive behavior.

Before installing, know that running the script creates or overwrites a local JSON briefing file, and optional cron setup can generate briefings on a schedule. Avoid enabling Telegram or other delivery integrations unless you understand where your personal reflections and priorities may be sent.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README describes receiving briefings and evening prompts via Telegram, but it does not include any privacy or disclosure warning about sending personal daily reflections or priorities through an external messaging service. For a markdown file, this is a user-affecting behavior involving potentially sensitive personal data and should be explicitly disclosed.

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill documents writing output to a local JSON file and references a script that supports an --output path, but the manifest does not declare any tool scope or permission boundary for file writing. In an agent ecosystem, undeclared write capability weakens user awareness and policy enforcement, which can lead to unexpected file creation or overwriting if the skill is invoked automatically.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The description uses broad activation language such as helping when 'starting your day, planning tasks, or maintaining daily routines,' which makes the skill eligible in many ordinary conversations without tight trigger boundaries. Over-broad routing increases the chance the agent invokes a skill with file-writing behavior in contexts where the user did not clearly request it, causing unintended actions or data generation.

Missing User Warnings

Low
Confidence
89% confidence
Finding
The markdown shows a command that saves output to daily_briefing.json but does not clearly warn that running it writes to disk. While this is not inherently malicious, lack of disclosure can surprise users, especially in automated or scheduled contexts, and may result in unintended persistence of personal routine data.

Static analysis

No suspicious patterns detected.