ClawHub
Back to skill

Security audit

cash-flow-forecast

Security checks for vulnerabilities and agentic risk

Overview

Review before use: the supplied script does not calculate the promised forecast and silently replaces missing or invalid financial figures with zero.

Treat the generated report as an unvalidated input summary. Do not rely on it for payroll, tax or liquidity decisions until weekly calculations and input validation are implemented and checked. Keep exported financial data and reports private. The optional Windows application is unnecessary for this workflow and requires a separate trust decision.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied code is an input-summary report generator rather than the declared forecasting tool. Its run function parses five scalar fields and writes calculation.md, but performs no forecast arithmetic, scheduling, or aggregation. The CSV reader is defined but never used. Local file reading and report writing are supporting implementation details, and there is no evidence of bank connectivity, money movement, or data exfiltration.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_write.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The skill produces an input summary rather than the advertised 13-week cash flow forecast.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

The script outputs supplied values and formula text but never calculates closing cash, cash after payroll and debt payments, or a 13-week forecast. In a financial-planning skill, this is an output-integrity defect: users or downstream agents could mistake successful report generation for completed forecasting and overlook liquidity shortfalls, although no direct money movement occurs.

Content

No source excerpt is available for this finding.

Missing or invalid cash inputs become zero despite the stated prohibition on inventing a cash balance.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

Missing, empty, or malformed financial inputs silently become numeric zero, making unavailable data indistinguishable from a confirmed zero balance or obligation. This contradicts the report's own guardrail against inventing balances and could conceal missing payroll, tax, or debt outflows when the report is used for financial decisions or downstream calculations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.