ClawHub
Back to skill

Security audit

Earnings Call Processor

Security checks across malware telemetry and agentic risk

Overview

This skill processes user-selected earnings call files locally and prepares a report, with no hidden publishing or credential behavior found.

Install only if you are comfortable processing the selected earnings call audio and stock data on this machine. Use trusted installs of Whisper and pandas, choose a dedicated output directory, and review the generated Feishu content before sharing or publishing it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill documentation instructs users to run local Python scripts, invoke the Whisper CLI, read input files, write outputs under /tmp, and execute a shell command that inlines file content into another command. Those are real code-execution and file-access capabilities, but no permissions are declared, which weakens sandboxing, review, and user awareness. In this finance-oriented workflow, the combination of shell, file read, and file write is expected functionally, but still increases risk because it processes local data and generates publishable documents that could expose sensitive financial information if misused.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.