ClawHub

Tamper-Proof Audit Trails for Trading Bots

Security checks across malware telemetry and agentic risk

Overview

This appears to be a trading/audit guide rather than malware, but it has enough under-disclosed production API, credential, logging, and validation-risk issues to require Review before install.

Install only if you are prepared to review and modify the examples before use. Treat all API examples as potentially production-bound, use test credentials first, avoid raw event-payload logging, redact sensitive trading/account data in reports and webhooks, and do not rely on the sample validation code for compliance until signature and retention checks are actually implemented.

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 (3)

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The guide states that all examples use the sandbox and that no API key is required to get started, but later Python examples point to the production API and explicitly require an API key. This kind of inconsistency can cause operators to send real trading or audit data to a production service under false assumptions, increasing the risk of unintended data disclosure, billing, or compliance mistakes.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The validation function claims to verify Ed25519 signatures and retention metadata, but the implementation only checks event presence, timestamp typing, and chain verification. This creates a false sense of compliance assurance: operators may rely on reports that omit core integrity and retention checks, leaving tampering or retention-policy gaps undetected.

Ssd 3

Medium
Confidence
95% confidence
Finding
The guide encourages logging full exception details, method context, and raw event payloads, then exporting and redistributing those records in reports and webhook flows. In a trading environment, these fields can contain credentials, proprietary strategy data, personal data, account identifiers, or other regulated information, causing broad secondary disclosure through audit systems that are often widely shared.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal