ClawHub

Forecast Valuation

Security checks across malware telemetry and agentic risk

Overview

This skill appears to generate a local Excel valuation workbook, but it overstates its financial-modeling capabilities and stores entered API credentials in plaintext.

Install only if you treat the workbook as a demo/template, not decision-grade valuation output. Do not enter real API credentials unless you accept plaintext config.json storage, and independently verify all generated financial assumptions, comparables, sensitivity tables, and target prices before business or investment use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
97% confidence
Finding
The script markets itself as a professional financial forecasting and valuation builder, but the generated workbook is largely populated with fixed sample numbers and canned assumptions rather than company-specific calculations. In a valuation skill, this is dangerous because users may rely on materially misleading outputs for investment decisions, especially since the workbook structure and formulas make the result look authoritative and customized.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The sensitivity analysis sheet is presented as a valuation analysis, yet the code explicitly generates simulated placeholder values with a simple arithmetic formula instead of linking to the DCF model. This can mislead users into treating fabricated sensitivity outputs as model-derived risk analysis, undermining financial decision-making and review controls.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill includes an option to upload generated valuation files to Baidu Netdisk without clearly warning that financial models, assumptions, and possibly proprietary company data may leave the local environment. In a finance workflow, this can expose sensitive internal analysis or regulated business information to third-party cloud storage unexpectedly.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The script collects API credentials and stores them in a local config.json file in plaintext, without warning the user about local persistence or encouraging safer secret handling. This increases the risk of accidental disclosure through source control commits, backups, shared directories, or permissive filesystem access, especially in a finance-oriented skill that relies on external data providers.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal