Back to skill

Security audit

unit-economics-calculator

Security checks for vulnerabilities and agentic risk

Overview

The bundled workflow has limited local access, but its script does not perform the promised calculations and should not be relied on for financial decisions.

Wait for corrected calculations and a matching input template before relying on this as a calculator. Independently verify every reported value because missing or invalid inputs become zero. The optional unsigned Windows app is unnecessary for the local workflow and needs separate assessment; a matching checksum alone does not establish safety.

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 declared primary purpose is numerical calculation, but run() performs only input parsing and report generation. No arithmetic computes net order value, contribution margin, CAC, payback, or LTV-to-CAC. Printing a formula does not fulfill the promise to compute these metrics with the math shown. Local file reading and report writing are supporting implementation details, and no unrelated access or undeclared external activity is evident.

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 calculator outputs inputs and formula text without computing the promised unit-economics metrics.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

The script produces a calculation report containing only normalized inputs and formula text; it never computes contribution margin, CAC, CAC payback, or LTV/CAC. In a skill intended to support financial decisions and board presentations, this is an output-integrity failure that could cause users or downstream agents to mistake an incomplete report for completed analysis, although no direct code-execution or data-exfiltration risk is evident. Silently converting missing or invalid inputs to zero and referencing undefined formula inputs further undermines the report's reliability.

Content

No source excerpt is available for this finding.