ClawHub
Back to skill

Security audit

Feishu Invoice Approval

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Feishu/Lark invoice approval bot scaffold with meaningful approval gates, dry-run defaults, and documented handling of sensitive invoice data.

Install only in an environment where the Feishu bot is allowed to read invoice messages and create approvals. Keep dry-run enabled until testing is complete, restrict Feishu app permissions, configure an uploader allowlist if needed, protect the local data directory, and define a retention policy for invoice images, extracted JSON, and SQLite records.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
83% confidence
Finding
The extractor writes structured invoice data to a predictable JSON file on disk, and invoices commonly contain sensitive personal and financial data such as taxpayer IDs, names, dates, and amounts. If output_dir permissions, retention, or cleanup are not tightly controlled elsewhere, this creates a privacy and data-exposure risk through local file access, backups, logs, or leftover artifacts.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.