ClawHub

vedic-destiny

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Vedic chart analysis/reporting skill that may run bundled local parsers and save local reports, with no evidence of hidden network access, credential use, or destructive behavior.

Install only if you are comfortable with the agent processing birth/chart data and saving local report files. Review where report folders are created, avoid including unnecessary personal details in inputs, and remove generated HTML/metadata/section files when you no longer need them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill instructs the agent to create local report artifacts, maintain report directories, write metadata files, and build self-contained HTML output as part of the workflow. These persistent side effects are broader than simple conversational analysis and can leak sensitive birth data or derived personal information into local storage without a clear consent boundary.

Context-Inappropriate Capability

Medium
Confidence
90% confidence
Finding
The skill normalizes execution of local Python scripts during routine operation, including chart validation and parsing tasks. Even if the referenced scripts are legitimate, instructing code execution from a content file expands the attack surface and can enable unsafe local actions in environments where the skill content is treated as operational authority.

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill instructs the agent to proactively generate `report.html`, which is an output artifact not declared in the skill metadata and exceeds a normal conversational response boundary. This creates a file-generation side effect that could write sensitive user-derived content to disk, confuse downstream tooling about expected outputs, or be abused as a stepping stone for broader unauthorized file creation behavior.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal