ClawHub
Back to skill

Security audit

TCS Expense Claim Processor

Security checks for vulnerabilities and agentic risk

Overview

This skill processes uploaded expense receipts into claim PDFs and spreadsheets, with sensitive but expected file handling for that purpose.

Install only if you are comfortable letting it process the receipts and invoices you upload for expense claims. Review the generated PDFs and spreadsheet before submitting them, avoid live FX lookup if you do not need it, and be cautious opening spreadsheets built from untrusted receipt text.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger text is excessively broad and includes generic phrases like processing bills, summaries, currency conversion, and organizing receipts, plus an instruction to 'Always use this skill' even for a single receipt. That can cause the skill to activate in situations beyond clear reimbursement intent, leading to over-collection of uploaded files, unnecessary parsing of sensitive financial documents, and unintended file generation or workflow execution.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The skill performs writes to `/mnt/user-data/outputs/` and generates PDFs/XLSX files, but the user-facing description does not clearly disclose that it will create persistent output artifacts in user storage. This reduces informed consent and may surprise users by duplicating or restructuring sensitive receipt and travel data into new files that could be retained, shared, or later accessed unintentionally.

Static analysis

No suspicious patterns detected.