Back to skill

Security audit

auto-repair-invoice-explainer

Security checks for vulnerabilities and agentic risk

Overview

The bundled code performs limited local text checks, although it does not deliver the promised invoice breakdown.

Expect a text-checking report, not automatic invoice interpretation or validation of charges. Review customer-facing explanations manually, omit unnecessary personal details, and choose an output folder where replacing lint-report.md is acceptable. Assess the optional Windows app separately before running it; the bundled workflow does not require it.

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

Although the lint checks use invoice-related section names, the script only checks keyword presence and formatting. Its output is a lint report, not an explanation of the supplied invoice. This is a materially different primary purpose, rather than merely a supporting implementation detail. Reading the input file and writing a report are ordinary implementation operations and are not independently flagged.

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 lints invoice-related text instead of explaining an invoice.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The manifest promises a breakdown of an auto repair invoice with plain-language explanations of labor, parts, fees, and taxes. The implementation only checks for section keywords, line lengths, and banned phrases, then writes a lint report; it neither interprets invoice charges nor produces customer-facing explanations.

Content

No source excerpt is available for this finding.