Back to skill

Security audit

bookkeeping-cleanup

Security checks for vulnerabilities and agentic risk

Overview

The skill provides a local bookkeeping checklist with approval safeguards, but its text checker cannot verify that your books are correct.

Use this as checklist guidance, not proof that accounts reconcile or transactions are complete. Independently verify balances, supporting statements, and proposed journal entries. The optional Windows application is unnecessary for the bundled workflow and needs a separate security assessment; the supplied checksum alone does not establish its 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 (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The description promises substantive bookkeeping analysis, but the executed code performs only textual checks. Required section names are matched as case-insensitive substrings without examining their contents or financial correctness. The CSV-reading helper is unused, and no executed logic analyzes ledger transactions. Local input reading and report writing support the linting function and are not themselves suspicious.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
86% confidence
Finding

The supplied command appears to lint cleanup notes or checklist content, which is materially narrower than performing the advertised bookkeeping cleanup. No ledger input or reconciliation behavior is visible. The missing Python implementation limits certainty, but the supplied code does not substantiate the declared primary capabilities.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
81% confidence
Finding

The declared purpose promises substantive ledger analysis, while the supplied command appears to lint a checklist YAML file. This suggests a materially narrower primary purpose, not merely a supporting implementation detail. Confidence is limited because the invoked Python script is unavailable; its internal behavior cannot be verified.

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 manifest promises ledger reconciliation and identification of every uncategorized or duplicate entry, but this script only lints checklist text.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

At L088–L103, the script checks only for section-name substrings, line lengths, and banned phrases; it does not inspect ledger transactions, reconcile balances, or identify uncategorized or duplicate entries. It can therefore report 'No problems found' at L108 for a document containing the expected headings even when the underlying books have unresolved issues, providing no support for the manifest's promised accounting checks.

Content

No source excerpt is available for this finding.