ClawHub
Back to skill

Security audit

Stock Batting Zone

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches its stated stock-report purpose, but it should be reviewed because it reads a local OpenClaw auth code and runs another skill's push script from a hardcoded path.

Install only if you are comfortable with this skill using your local OpenClaw today-task auth code to push reports and with it depending on a hardcoded local today-task script path. Review or adjust the credential source and push script path before enabling scheduled unattended runs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
The skill reads an auth code from a separate local configuration file belonging to another skill path, which is a credential-access behavior not obviously required for stock analysis. This creates unnecessary secret coupling and can expose or misuse credentials without clear user consent or scope boundaries.

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
The script launches an external Python script to perform the push operation, giving this skill broader execution capability than simple report generation implies. Even if intended for modularity, invoking external code increases trust boundaries and lets behavior change independently of this script's reviewed logic.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
A credential-like auth code is silently loaded from local configuration and embedded into the outbound payload for pushing content, with no user warning, consent, or explicit scope control in the script. In the context of a stock-reporting skill, hidden credential use is more suspicious because the core function does not inherently require cross-skill secret harvesting.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.