ClawHub
Back to skill

Security audit

Ai Drama Prompt Factory.Skip

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent short-drama prompt generator, but its bundled document tools can broadly index local files beyond clearly scoped story materials.

Install only if you will use it in a dedicated project folder containing the novel or story files you intend to process. Do not point its indexing tools at your home directory, downloads, workspace root, source repositories, or folders that may contain secrets, credentials, or private business files. Treat generated index files as sensitive and delete them when they are no longer needed.

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 (2)

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
This script provides a general-purpose document/code indexing capability that is unrelated to the stated short-drama prompt-factory purpose. In an agent skill, unjustified file-enumeration and content-indexing functionality expands the attack surface and can enable unintended collection of local source code, documents, metadata, and keywords from the host workspace.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The directory indexing logic recursively scans many code and document formats, computes hashes, records absolute paths, timestamps, file sizes, structure, and extracted keywords, then aggregates them into a JSON index. In the context of a short-drama generation skill, this is over-privileged behavior that could expose sensitive project contents or environment information if run on a broad path such as the repository root or home directory.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.