Back to skill

Security audit

solar-quote-verifier

Security checks for vulnerabilities and agentic risk

Overview

This offline solar calculator has proportionate access, but confirmed calculation and reporting errors mean its financial conclusions need independent checking.

Use this as a rough comparison aid. Independently recalculate the 25-year return, verify current tax-credit eligibility and utility export rates, and read every warning before relying on a FAIR label or signing a contract.

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 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

The documented 25-year net calculation actually returns a 30-year total.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

payback() accumulates 30 years of cash flows by default, but its returned cumulative value is displayed and evaluated as the 25-year net. This can overstate the stated 25-year financial outcome and suppress the negative-net warning for proposals that only become profitable in years 26–30, undermining the skill's core verification purpose.

Content

No source excerpt is available for this finding.

ITC eligibility is assumed rather than verified as promised by the manifest.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The script treats users as eligible for a 30% ITC unless they explicitly supply --no-itc; it does not verify eligibility or the applicable credit rate. Although the output discloses an assumption, the skill promises eligibility verification, so users may rely on understated costs and overly favorable payback estimates when making a substantial purchase.

Content

No source excerpt is available for this finding.

Detected red flags can still produce a FAIR final verdict.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

When roof age, negative cumulative returns, or an unrealistic quoted payback sets ok to false, the final output can still use price_v, which may remain 'FAIR (market range)'. The loan cash-flow warning also leaves ok unchanged, allowing a FAIR verdict despite that warning; users or downstream agents relying on the headline can consequently miss material financial risks.

Content

No source excerpt is available for this finding.