ClawHub
Back to skill

Security audit

Invoices

Security checks across malware telemetry and agentic risk

Overview

This invoice-management skill stores accounting records locally, and the persistent writes are disclosed, purpose-aligned, and bounded to the stated Clawic data folders.

Before installing, be comfortable with the agent maintaining a local invoice archive and accounting memory under ~/Clawic/data, including supplier tax IDs, invoice history, open items, and last-four bank details. Review the recurring Due table and shared finance/contact/project boxes if you want tighter control over what gets updated automatically.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Low
Confidence
90% confidence
Finding
The skill instructs the agent to write newly seen suppliers, corrected tax IDs, and parsing quirks into persistent memory files, but this document does not present an explicit user-facing warning or consent boundary for that persistence. In an invoice-processing context, those writes can contain sensitive business and tax data, so silent persistence increases privacy and data-governance risk even if the feature is operationally useful.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill instructs the agent to perform recurring outputs automatically and to write back to persistent artifacts like `memory.md`, `## Due`, `## Open Items`, and accountant handoff files without requiring a fresh user confirmation at execution time. In an agent setting, this creates a state-changing automation path where periodic runs can modify records, generate external-facing deliverables, and overwrite operational history based only on prior configuration, increasing the risk of unintended file changes, stale assumptions, or unauthorized propagation of sensitive accounting data.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly instructs the agent to make persistent repository changes ('goes to ## Suppliers' and 'goes to ## Open Items') as part of normal search behavior, but it does not require explicit user confirmation before modifying stored data. In an archive/search skill, this creates a real integrity risk: a simple query can silently alter business records or memory files, causing unauthorized or incorrect persistent updates.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
memory-template.md:68