ClawHub
Back to skill

Security audit

AR Collections Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AR collections helper that drafts reports and messages while requiring approval for external sends and accounting changes.

Before installing, make sure your organization replaces the named approver with its own authorized role or person, keeps client AR data separated, and reviews all customer-facing collection messages and any QBO/accounting changes before execution.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
---
name: ar-collections-agent
description: Accounts Receivable (AR) collections workflow automation for accounting firms and finance teams. Use when you need to: (1) identify overdue invoices and aged AR buckets (30/60/90/120+ days), (2) draft escalating collection emails or call scripts, (3) generate AR aging reports and DSO calculations, (4) prioritize collection targets by balance and risk, (5) track payment promises and follow-up schedules, (6) calculate bad debt reserves or write-off recommendations, or (7) produce client-ready AR health dashboards. Works with QBO exports, CSV invoice data, or direct QBO API. NOT for: initiating legal action or filing liens (escalate to attorney), sending external emails without approval, accessing client QBO accounts without explicit write authorization, or PTIN-backed tax services.
---

# AR Collections Agent
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The skill hard-codes approval to a specific individual ('Irfan') instead of using a role-based or configurable approver model. This creates a governance and availability risk: if the named person is unavailable, replaced, or inappropriate for a given deployment, users may bypass controls or apply the skill incorrectly.

Static analysis

No suspicious patterns detected.