ClawHub
Back to skill

Security audit

Receipt Auditor

Security checks across malware telemetry and agentic risk

Overview

The skill does what it claims, but its CLI silently saves users' receipt and bill contents locally despite claiming no data storage.

Review before installing if you may process real receipts, credit-card charges, reimbursements, or travel details. Use only redacted test data unless you are comfortable with the skill writing the full prompt to a local last-prompt.txt file, and consider removing or changing that save behavior before using it with sensitive financial information.

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 (2)

Missing User Warnings

Medium
Confidence
85% confidence
Finding
The skill is designed to process receipts, bills, expense lines, merchants, travel history, and other potentially sensitive financial data, but it does not give users a clear upfront warning about privacy risks, data minimization, or handling expectations. Even though it claims 'Privacy-first: No data storage — process and forget,' that assurance is buried later and is not presented as an explicit consent or caution point before users provide sensitive inputs.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The script persistently writes the full generated prompt to references/last-prompt.txt, including receipt contents from CLI arguments or files. Receipts and bills often contain sensitive financial or personal data, so storing them without explicit consent, warning, or access controls creates a local data exposure risk, especially on shared systems or in synced working directories.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.