Back to skill

Security audit

cash-flow-forecast

Security checks for vulnerabilities and agentic risk

Overview

The skill promises a working cash forecast, but its script performs no forecast calculations and silently replaces missing or invalid amounts with zero.

Do not rely on this report for payroll, tax or funding decisions until weekly calculations and input validation are implemented and independently checked. Keep financial exports private. The optional unsigned Windows app is unnecessary for the Python workflow and requires separate review.

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
  • 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 (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The static analysis reports that the implementation substitutes echoed inputs and descriptive text for the promised weekly cash calculations. If accurate, this is an output-integrity defect: users could mistake an incomplete report for a functioning forecast and miss payroll or tax funding shortfalls, although the script was not supplied for independent verification. The financial decision-making context increases the consequences, while the disclaimer against guaranteed results does not resolve missing calculations.

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 outputs supplied values and a formula string instead of building the promised 13-week cash flow forecast.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

The script parses and repeats five supplied values and prints a formula, but never evaluates that formula or constructs the promised 13-week forecast. In a cash-management workflow, this functionality and output-integrity failure could cause users or downstream agents to mistake an incomplete report for a completed forecast and miss liquidity shortfalls; the printed guardrails do not compensate for the missing calculations.

Content

No source excerpt is available for this finding.