ClawHub
Back to skill

Security audit

Agent Org Manager

Security checks for vulnerabilities and agentic risk

Overview

The available evidence points to a purpose-aligned multi-agent automation skill, with user-visible caution needed around optional cron templates that may write reports and share summaries across sessions.

Before installing or enabling scheduled jobs, review the cron templates and confirm where they write files and which session receives reports. Keep summaries free of secrets, customer data, and private prompts unless you explicitly want that information shared across sessions.

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 (2)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The template instructs an automated cron-driven agent to write files into the repository (`company/departments/[dept]/learning_YYYYMMDD.md`) without any disclosure, approval gate, or path restriction guidance. In an agent-org skill, this creates a real integrity risk because scheduled background runs can silently modify project data and accumulate unreviewed content.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The daily report template directs automated transmission of work summaries to another session (`sessions_send to main`) without any privacy notice, scope limitation, or approval control. In this skill's multi-agent organizational context, that increases the chance of silent cross-session sharing of sensitive operational data or prompts beyond what a user expects.

Static analysis

No suspicious patterns detected.