Back to skill

Security audit

Financial Close Checklist

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed financial-close checklist skill with no executable code, persistence, credential access, or hidden data movement.

Before using this with live accounting integrations, confirm the client/entity, period, and reviewer, and require explicit approval before posting entries, locking periods, or distributing financial reports. As packaged here, it is guidance-only Markdown rather than an automation that can act on accounts by itself.

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
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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill explicitly instructs locking accounting periods, exporting financial statements, and distributing packages, which are impactful actions in a financial system. Without an explicit confirmation gate or warning that these steps modify records/access controls and may affect downstream reporting, an agent could perform sensitive changes prematurely or on the wrong period/entity.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
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.

Content

Scanner excerpt · SKILL.md (reported line 350)May include surrounding context.

text
1. Automate bank feeds — eliminate manual download/import
2. Set transaction cutoff rules — no backdating without approval
3. Standardize chart of accounts — fewer coding questions = faster entry
4. Create recurring JE templates — one-click accruals, depreciation, amortization
5. Use close management software — FloQast, Blackline, or a simple Google Sheet tracker

Static analysis

No suspicious patterns detected.