Back to skill

Security audit

Copilot Money Mac

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local, read-only helper for analyzing Copilot Money data, with sensitive access that matches its stated purpose.

Install only if you are comfortable letting the agent read and summarize your local Copilot Money financial records. Prefer specific questions and avoid asking it to dump raw database or cache contents unless you really need that detail.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is designed to access and analyze highly sensitive financial records, including transactions, balances, budgets, and investments, but provides no privacy disclosure, consent guidance, or data-handling limitations. This increases the risk of over-collection, unintended exposure, or analysis of sensitive information without clear user awareness or minimization boundaries.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill explicitly instructs use of shell pipelines (strings, grep, sort) against local Firestore LevelDB files, which expands the skill from structured finance queries into arbitrary file-scanning and subprocess-style data extraction. In a skill handling personal finance data, this is dangerous because it normalizes unstructured scraping of highly sensitive local data stores and can expose more information than needed for the user’s request.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.